Upload
others
View
1
Download
0
Embed Size (px)
Citation preview
SOIL TEST BASED INTEGRATED NUTRIENT
TAILORING FOR OPTIMUM BANANA PRODUCTION
AND SUSTAINABLE SOIL HEALTH USING ARTIFICAL
NEURAL NETWORKS
Thesis submitted in
Partial Fulfillment for the award of
Degree of Doctor of Philosophy
In Computer Science
By
N.MANOHARAN,(Reg.No.M698800014)
Supervisor
Dr.R.BALASUBRAMANIAN,Ph.D., M.Phil (Maths)., M.Phil (C.S.)., M.Phil (Mgt)., M.S., M.B.A., M.A.D.E.,
PGDIM., PGDOM., PGDCA., PGDHE., DIM., DDE., CCP.,
Professor & DeanFaculty of Computer Applications
Erode Builder Education Trust’s Group of InstitutionsNathakadaiyur, Kangayam.
VINAYAGA MISSION’S UNIVERSITY,
SALEM-636 308, Tamil Nadu, India.
June-2012
DECALARATION
DECLARATION
I, N.MANOHARAN declare that the thesis entitled “SOIL TEST BASED
INTEGRATED NUTRIENT TAILORING FOR OPTIMUM BANANA
PRODUCTION AND SUSTAINABLE SOIL HEALTH USING ARTIFICAL
NEURAL NETWORKS” submitted by me for the Degree of Doctor of Philosophy
in Computer Science is the record work carried out by me during the period from
2006 to 2012 under the guidance of Dr. R.BALASUBRAMANIAN, and has not
formed the basis for the award of any degree, diploma ,associate-ship, fellowship,
titles in this University or any other University or other similar institutions of higher
learning.
Signature of the Candidate
Place:
Date:
CERTIFICATE
VINAYAKA MISSIONS UNIVERSITY
CERTIFICATE BY THE GUIDE
I, Dr. R.BALASUBRAMANIAN certify that the thesis entitled “ SOIL TEST
BASED INTEGRATED NUTRIENT TAILORING FOR OPTIMUM BANANA
PRODUCTION AND SUSTAINABLE SOIL HEALTH USING ARTIFICAL
NEURAL NETWORKS ” submitted for the Degree of Doctor of Philosophy in
Computer Science by Mr. N.MANOHARAN, is the record of research work carried
out by him/her during the period from 2006 to2012 under my guidance and
supervision and that this work has not formed the basis for the award of any degree,
diploma, associate-ship, fellowship or other titles in this University or any other
University or Institution of higher learning.
Place: Signature of the Supervisor with designation
Date:
ACKNOWLEDGEMENT
i
ACKNOWLEDGEMENT
I would like to thank my guide Dr. R. Balasubramanian, Dean, Computer
Applications, ETBT Group of Institutions, Erode for his inspiring and stimulating
guidance, insightful critical comments and warm encouragement. I learned many
important things of being a good researcher from him such as being patient,
persistence, remaining modest and prudent, keeping open mindedness, as well as
being positive and keeping practising academic writing.
My sincere thanks goes to Dr. C.Pooranachandran , Head of the
Department of Computer ,Government Arts College, Nandanam, for their advice,
support and help during my this study.
My sincere thanks goes to Dr. K.J.Jeyabaskaran ,Scientist-Soil Science,
National Research Centre for Banana, Trichy, for their advice, support and help
during my this study.
I would like to thank , the principal and all faculty members and supportive
staff members of Department of Computer Science, HH The Rajah’s College,
Pudukkottai who extended their full support right from my enrolment to till date
enabling me to carry out my research work successfully.
During the course of my studies, there are various institutions which provided
work opportunities, learning materials extended intellectual support and allowed to
me collect necessary data. In particular, I am very much grateful to the institutions,
National Research Centre for Banana (NRCB), Trichy.
I am also indebted to all the respondents and students for their support and
active participation in the phases of data collection and evaluation.
ii
I would like to thank my fellow research scholars with whom I shared
friendship, ideas and much debate during my time.
My special thanks to my wife for being there and supporting me patiently
from her work place during my research work. She shared her experience in IT that
helped me a lot during the development of the frame work and working version of the
system.
I also owe many thanks to the participants who had voluntarily shared their
experiences and perspectives with me in the interviews for this study.
Many people have contributed in diverse ways to my studies that lead to the
completion of this work, including the staff and colleagues in the Department of
Computer Science, SRM, Arts Science College, Chennai and National Research
Centre for Banana (NRCB), Trichy given their time to listen my ideas, shared their
experience, reviewed my writing and provided significant suggestions.
I would like to thank my friend, Mr. Selvakumar , Assistant professor,
Department of Computer Science, SRM, Arts Science College, and his wife
V.Karthiga Devi, Assistant professor, Department of English, Queen Mary’s College,
ideas and shared their experience, reviewed my writing and provided significant
suggestions.
More importantly I would like to thank my father, mother, sister, brother,
daughter and the almighty for showing me the right directions out of the blue and help
me to stay calm in the oddest of times and keep moving even at times when there was
no hope.
MANOHARAN.N
CONTENTS
iii
CONTENTS
ACKNOWLEDGEMENT i
LIST OF TABLES vii
LIST OF FIGURES viii
EXPANSION/DESCRIPTION ix
Abstract xi
CHAPTER I - INTRODUTION 1
1. Introduction 1
1.2 Research Motivation 8
1.3 Soil and factors 9
1.3.1 Soils 9
1.3.2 Climatic and topographic factors 9
1.3.3 Effects in the banana root system 9
1.3.4 Climate and topography 9
1.3.5 Biological factors 10
1.4 Research Problem 11
1.5 Aim and Objectives 12
1.6 Scope of the Study 13
1.7 Limitations of the study 14
1.8 Significance of the research 15
1.9 Statement of Problem 16
1.10 Structure of the Thesis 17
iv
CHAPTER II- LITERATURE SURVEY 18
2.1 Soil based banana plant yield 18
2.2 Artificial Neural Networks based 21
Other crops yield methods
2.3 Applications of Fertility Gradient approach for 23
various crop yield prediction
2.4 Costs and returns of tissue culture banana and 24
sucker propagated banana
2.5 Resource use efficiency in banana production 29
2.6 Marketing channels and marketing costs 32
2.7 Problems in production and Marketing of banana 36
2.8 Synthesis of the Literature Review 40
CHAPTER III-SURVEY ANALYSIS AND DESIGN 41
3.1 Introduction 41
3.2 Soils 41
3.3 Major soil types of India 43
3.3.1 Red Soil 43
3.3.2 Lateritic soil 43
3.3.3 Black soil 43
3.3.4 Alluvial soils 43
3.3.5 Desert soils 43
3.3.6 Forest and Hill Soils 43
high in organic matter
3.4 Soil Groups 43
3.4.1 Red Soils 44
v
3.4.2 Laterite and lateritic soils 46
3.4.3 Black soils 48
3.4.4 Alluvial soils 50
3.4.5 Desert Soil 51
3.4.6 Forest and hill soils 51
3.5 Plant Analysis as Nutritional 52
Requirements of Bananas
3.5.1 Stage of sampling. 58
3.5.2 Taking representative sample 59
3.5.3 Plant Analysis Interpretation 59
3.6 Artificial Neural networks and its Applications 65
3.6.1 Use of neural networks 65
3.6.2 Advantages 65
3.6.3 A simple neuron 66
3.6.4 Sophisticated Neuron 66
3.6.5 Applications 68
3.7 Summary 70
CHAPTER IV- METHODOLOGY 71
4.1 Introduction 71
4.2 Nutrient requirement of a banana crop 72
4.3 Absolute Update Technique for 73
ANN Banana Yield Prediction
4.3.1 Structure of Absolute Update Technique 73
4.3.2 Architecture of Absolute Update Technique Design 75
4.4 Soil Analysis 78
4.4.1 Data Source 78
vi
4.4.2 Creation of fertility gradient in the soil of 80
experimental field
4.4.3 The selected Treatment Plots in the Combination 82
4.5 Leaf Analysis 83
4.6 Summary 86
CHAPTER V - RESULT AND DISCUSSION 87
5.1 Introduction 87
5.2 Absolute Update Technique using soil test 88
5.3 Absolute Update Technique using Leaf Nutrients 98
5.4 Discussion 105
CHAPTER VI- 6 COMPREHENSIVE CONCLUSIONS 111
AND SCOPE OF THE FUTURE WORK
6.1 Summary of the Present Work 111
6.2 Future work 112
REFERENCES 113
LIST OF PAPERS PRESENTED/PUBLISHED 126
APPENDIX
Appendix-I Glossary and Terms 128
LIST OF TABLES
vii
LIST OF TABLES
Table No Title Page No
1 Critical levels of nutrients in different tissue of
completely developed banana plants
64
2 Treatment combinations 82
3 The initial soil test values available N P K gram per
plant
88
4 Need for NPK Nutrients for kg per hectare 90
5 N P K Available in the Fertilizers 92
6 Fertilizer ratio in plant 94
7 Expenses and Target and Profit 96
8 Leaf Nutrients based combination -1 98
9 Leaf Nutrients based Combination -2 100
10 Leaf Nutrients based Combination -3 102
11 Final yield prediction of Leaf Nutrients 104
LIST OF FIGURES
viii
LIST OF FIGURES
Figure
No Title Page No
1 Sampling procedure for banana leaves 54
2 The spatial variability in the mineral content of leaf
blade of banana cultivar Drawf Canvendish
57
3 Relationship between essential nutrient connection
and plant growth or Yield
60
4 Concentration of Nutrient in tissue 61
5 A simple neuron 66
6 MCP neuron 67
7 Architecture of Artificial Neural Networks 67
8 The prototype model of banana yield prediction 72
9 Absolute Update Technique based ANN 74
10 Absolute Update Technique Architecture 76
11 Morrow plots taken in the experiment field 78
12 Morrow plots taken in the experiment field has split
up into four trips
79
13 Morrow plots soil experimental field 80
14 The sampling procedure of banana leaf 85
15 The nutrients levels in the banana plant 106
16 The effect of nutrients levels of NPK on Nendran
banana plant
107
17 The effect of nutrients levels of NPK on Rasthali 108
18 The effect of nutrients levels of NPK on banana plant 109
19 The effect of nutrients levels of NPK on banana plant
on Nutrient Tailoring
100
ix
EXPANSION AND DESCRIPTION
1. ANN Artificial Neural Networks
2. GRT green revolution technologies
3. CNR Critical Nutrient Range
4. DRIS Diagnosis and Recommendation Integrated System
5. PASS Plant Analysis with Standardized Scores (PASS),
6. CNC Critical Nutrient Concentration
7. CDL Critical Deficient Level
8. CTL the Critical Toxic Level
9. NRCB National Research Centre for Banana
10. MLR multiple linear regression
11. SMLR stepwise MLR
12. PLSR partial least squares regression
13. PPR projection pursuit regression
14. BPN Back-propagation neural network
15. Sigma Six Sigma is a management philosophy developed by Motorola
16. Acrylic Expression Graphic Designer is Microsoft's
17. ACT Application Compatibility Toolkit (ACT)
18. ASE Adaptive Server Enterprise
19. APO Advanced Planner and Optimizer
20. APS Advanced planning and scheduling
21. ASD Agile software development
x
22. AI Artificial intelligence
23. Ajax Ajax (Asynchronous JavaScript and XML)
24. ALPR Automated License Plate Recognition
25. CRM customer relationship management
26. AP An artificial passenger
29. APM Application portfolio management
30. APO Advanced Planner and Optimize
31. APPC Advanced Program-to-Program Communication
32. ATS Applicant tracking system
33. APS Advanced planning and scheduling
34. AR Augmented reality
35. ARAX Asynchronous Ruby and XML
36. ASE Adaptive Server Enterprise
37. ASP An application service provider
ABSTRACT
xi
Abstract
Setting a realistic yield goal in each part of the field is one of the critical
problems in precision for agriculture. The factors affecting crop yields, such as soil,
weather, and land management. So complex and traditional statistical methods do not
give accurate results. Once yield goals are set the correct amount of seed, fertilizer,
etc., to produce target yields. In Crop yield history suggests that crop production
systems are very complex. If less fertilizer is applied, the yield may be reduced, if too
much is applied, money will be wasted and the environment may suffer.
Banana production systems at the current level of yields are not found to be
sustainable, in the long run, as there is significant reduction of plant nutrients in soil.
An Artificial Neural Network was used to build a crop yield prediction model for
precision farming applications.
ANN as a base, development of new model was used to be very useful in
setting more realistic target yields within fields for precision agriculture. Specifically
a Artificial neural network’s ability to predict various banana indices was tested and
its accuracy was compared against a Fertility Gradient approach and linear methods,
existing statistical method, as well as the neural network method with back
propagation algorithm.
The newly developed model was Absolute Update Technique involved for
various analyses and reduces the more number of iterations. The Absolute Update
Technique has been used to drastically reduce the dimension of the network and
computational effort. The Absolute Update Technique is mainly based on inorganic
fertilizers and integration of organic nutrient sources based on initial soil test values
and leaf nutrients through the development of farmer-friendly and based on the
financial position of the farmers, without affecting the soil health and farmers’ wealth
adversely.
1
SOIL TEST BASED INTEGRATED NUTRIENT TAILORING
FOR OPTIMUM BANANA PRODUCTION AND SUSTAINABLE
SOIL HEALTH USING ARTIFICAL NEURAL NETWORKS
CHAPTER -I1. Introduction
Banana (Musa)1, a plant genus of extraordinary significance to
human societies, produces the fourth most important food in the world
today (after rice, wheat, and maize), bananas and plantains. Banana species grow
in a wide range of environments and have varied human uses, ranging from
the edible bananas and plantains of the tropics to cold-hardy fiber and
ornamental plants. The plant is a source of food, beverages, fermentable
sugars, medicines, flavorings, cooked foods, silage, fragrance, rope,
cordage, garlands, shelter, clothing, smoking material, and numerous
ceremonial and religious uses. With the exception of India, banana and
plantain are ideally suited for traditional and agro forestry, for inter planting
in diversified systems, and for plantation-style cultivation in full sun.
Although mostly consumed locally in the Indian region, the fruit enjoys a
significant worldwide export market.
Indian agriculture has responsibility of providing national as well as
household food and nutritional security to its teeming millions in a scenario of
planting genetic potential in all major crops and declining productivity in vast tracts
of rain fed/ dry land areas constituting approximately 44.2 percent of net cultivated
area. Wide-spread occurrence of ill-effects of green revolution technologies (GRTs)2
in all intensively cultivated areas is threatening the very sustainability of the
important agricultural production systems and national food security. It has also to
share local as well as global responsibilities to ensure environmental safety for human
kind.
1 Biological name of Banana is musa2 Green Revolution Technologies is used to share the global and local responsible for environmentsafety for human kind
2
A mismatch between the national food grain production and requirement has
already crept into the system, which is further widening. The human population of
India has increased to 1210.2 million at a growth rate of 1.76 per cent in 2011 over
2001 (1028.7 million) and is estimated to increase further to 1530 million by 20303
(Census of India, 2011). On the other hand our national food grain production for past
3-4 years is hovering around 234 million tonnes. This means that per capita food
grain production is only about 193 kg per year. There are projections that demand for
food grains would increase from 234 million tonnes in 2009-10 to 345 million tonnes
in 2030 (GOI, 2009)4. Hence in the next 20 years, production of food grains needs to
be increased at the rate of 5.5 million tonnes annually [99].
Simultaneously, the demand for high-value commodities such as fruits,
vegetables, livestock products, fish, poultry etc., is increasing faster than food grains,
and is expected to increase by more than 100% from 2000 to 2030. As a result, area
under horticultural crops has increased appreciably during past two decades .At
present, more than 20 million hectare area is reported under horticultural crops with a
total production of 207 million tonnes, of which major contribution comes from fruits
(60.8%) and vegetables (30.7%). The fruits are grown in approximately 5.78 million
hectare with a production level of 63.50 million tonnes. Likewise, total production of
vegetables is about 125.90 million tonnes which comes from an area of 7.80 million
hectare (Agricultural Situation in India, 2009) Of the total vegetable production, more
than 65 percent comes from potato, tomato, onion, brinjal, okra, cabbage and
cauliflower.
3 Census of India 2011 to increase food grain production and requirement for increasing population inindia.
4 Vision 2030 Project Director, Project Directorate for Farming Systems Research (ICAR),Modipuram,Meerut-250 110 (U.P.), India. Typeset & Printed in: Yugantar Prakashan Pvt. Ltd.,WH-23, MayapuriIndustrial, Area, Phase-I, New Delhi.
3
From an historical point of view, the understanding of the banana soil-plant
relationship can be divided into periods before and after 1990. Before 1990, the
banana industry in general was very respectful of soil quality; only soils with
optimum morphological, physical and chemical conditions were placed under banana
production.
It can be stated without a doubt that climate and soil determine the success of
banana production enterprises. In most cases, climatic factors are easier to determine;
however, the soil component is much more difficult to characterize due to the
variation of the soil morphological, physical and chemical properties within a given
area (large or small) under the same climate. The effect on banana root performance
of the various components of these two factors has not been fully understood mainly
due to the interactions that occur among them. However, the effect of some soil
properties on root performance has been understood to a considerable extent after the
experience gained during the great expansion of the banana industry in the 1990s.
Plant analysis has been considered a very practical approach for diagnosing
nutritional disorders and formulating fertilizer recommendations (Kelling et al.,2000;
Self, 2005) [49]. Plant analysis, in conjunction with soil testing, becomes a highly
useful tool not only in diagnosing the nutritional status but also an aid in management
decisions for improving the crop nutrition (Rashid, 2005) [80].
Plant analysis is the quantitative analysis of the total nutrient content in a plant
tissue, based on the principle that the amount of a nutrient in diagnostic plant parts
indicates the soil’s ability to supply that nutrient and is directly related to the
available nutrient status in the soil (Malavolta, 1994 [62]; Kelling et al., 2000 [49];
Havlin et al., 2004[37]; Rashid, 2005[80]). It is a very practical and useful technique
for fruit trees and long duration crops (Rashid, 2005) [80]. Hence, it seems quite
convenient and appealing for bananas also.
4
Bananas are heavy feeder of nutrients (Jones, 1998) [47] and thus need balanced
nutrition for optimum growth and fruit production, and gives in turn potential yields.
A deficiency or excess of nutrients can cause substantial damage to the plant (Memon
et al., 2001) [68]. The early (until the mid-1960s) researches on banana nutrition had
concentrated on the description of symptoms of nutrient imbalance and the conduct of
field experiments comparing response to rates of applied fertilizer on a range of soil
types. During last three decades, scientists attempted to understand more clearly the
role of nutrients in the growth and development of bananas. Field studies of fertilizer
response are still being conducted, but attempts to relate nutrient concentrations in the
soil and plant to yield have complemented this work. Analysis of plant parts for
mineral elements and the attempt to set standards for interpreting leaf analysis data
came to the fore in the late 1960s and early 1970s.
However, each researcher approached the problem differently, probably
reflecting a lack of unifying concepts in the understanding of the growth and nutrition
of bananas, until Martin-Prevel (1974 [64], 1977 [65]) initiated the formation of an
International Group on Mineral Nutrition of the Banana that resulted in a suggested
International Reference Method for sampling in banana fertilizer experiments.
Plant analysis can serve as a nutritional guide. Plant analysis, normally, is a
laboratory analysis of collected plant tissue. Using established critical or standard
values, or sufficiency range, a comparison is made between the laboratory analysis
results with one or more of these known values or ranges in order to access the plant’s
nutritional status (Jones et al., 1991 [45]; Kelling et al., 2000 [49]; Rashid, 2005
[80]). Hence, it can be successfully used to identify the hidden hungers of plants (PPI,
1997; Kelling et al., 2000; Tisdale et al., 2002; Rashid, 2005).
The use of plant analysis as a diagnostic tool has a history dating back to
studies of plant ash content in the early 1800's. While working on the composition of
plant ash, researchers recognized the existing relationships between yield and the
nutrient concentrations in plant tissues. Quantitative methods for interpreting these
5
relationships in a manner that could be used for assessing plant nutrient status arose
from the work of Macy (1936) [60]. Since then, much effort has been directed
towards plant analysis as diagnostic tool.
Banana is one of the important fruit crops in India. In India, 16.5 million
tones of banana are being produced from 4.5 lakh hectares, with 16-lakh tones of
inorganic fertilizers, annually. By 2020, India has to produce 25 million tones of
banana for exploding population and for export purposes.
For this requirement of inorganic fertilizers it is extrapolated to about 25 lakh
tones. The cost of inorganic fertilizers is increasing day by day. Fortunately, N and P
fertilizers are manufactured / mined in India sufficiently based on our needs and their
prices are in our control, but unfortunately, we are depending on foreign countries for
K fertilizers, which account nearly 50 per cent of the cost of inorganic fertilizers
required for banana. There is no chance of reduction in hike in cost of K fertilizers
and hence in the cost of total fertilizer input for banana. As this is the present
situation, the target, 25 million tones of banana is likely to be out of reach.
On the other hand, banana being a K-loving crop depletes soil K rapidly and
replacement of K in banana soils is not in proportion due to increasing cost of K
fertilizers in the market and hesitation of the banana farmers to application of such a
costly K fertilizers adequately to their soils. Such type of problems leads to sever
nutritional imbalances, which are the permanent soil health damages.
Crop yield history suggests that crop production systems are very complex.
Both process-oriented crop growth models and traditional statistical methods can be
used to study crop growth and yield response to environment and management. For
example, Paz et al. (1999, 1998) [76] developed a technique to characterize corn yield
variability using the CERES-Maize process-oriented crop growth model, and to
characterize soybean yield variability using the CROPGRO-Soybean process-oriented
crop growth model.
6
Drummond et al. (1995) [23] compared several methods for predicting crop
yield based on soil properties. They noted that the process of understanding yield
variability is made extremely difficult by the number of factors that affect yield. They
used several multiple linear regression methods --- such as multiple linear regression
(MLR), R 2 = 0.42; stepwise MLR (SMLR), R 2 = 0.43; partial least squares
regression (PLSR), R 2 = 0.43; projection pursuit regression (PPR), R 2 = 0.73; and
back-propagation neural network (BPN), R 2 = 0.67 --- for modeling the relationship
between corn yield or soybean yield and soil properties. They concluded that less-
complex statistical methods, such as standard correlation matrices, did not seem to be
particularly useful in understanding yield variability. The correlation matrices
described each factor's linear relationship to yield. However, when complex nonlinear
relationships between factors exist, correlation may provide inaccurate and even
misleading information about these relationships.
Data mining tools are beginning to show value in analyzing massive data sets
from complicated systems and providing high-quality information (White and Frank,
2000) [97]. An artificial neural network (ANN) is an attractive alternative for
building a knowledge-discovery environment for a crop production system.
Ambuel et al. (1994)[3] used a “fuzzy logic expert system” to predict corn
yields with promising results. The functional relationship using the fuzzy logic expert
system was expressed linguistically instead of mathematically. The authors suggested
the use of a neural network to predict within-field yields.
Mining of nutrients from soil is a major problem causing soil degradation and
threatening long-term food production in developing countries. In present research, an
attempt was made for carrying out nutrient budgeting, which includes the calculation
of nutrient balance at (plot / field) and meso (farm) level and evaluation of trends in
nutrient mining / enrichment.
7
Banana production systems at the current level of yields are not found to be
sustainable, in the long run, as there is significant reduction of plant nutrients in soil.
Hence Artificial Neural Network can be used to build a crop yield prediction model
for precision farming applications.
Setting a realistic yield goal in each part of the field is one of the critical
problems in precision for agriculture. The factors affecting crop yields are soil,
weather, and land management. Traditional statistical methods do not give accurate
results. Crop yield history suggests that crop production systems are very complex. If
less fertilizer is applied, the yield may be reduced, if too much is applied, money will
be wasted and the environment may suffer.
ANN as a base, development of new model was practised and found to be
very useful in setting more realistic target yields within fields for precision
agriculture. The newly developed model is an Absolute Update Technique, which
involves various analyses and reduces the more number of iterations. The Absolute
Update Technique has been used to drastically reduce the dimension of the network
and computational effort. It is mainly based on inorganic fertilizers and integration of
organic nutrient sources based on initial soil test values. This method is very
economical that is very farmer-friendly and can be easily practised by them without
affecting the soil health and farmers’ wealth.
8
1.2 Research Motivation
Banana is one of the important fruit crops in India. In India, 16.5 million tones
of banana are being produced from 4.5 lakh hectares, with the 16-lakh tones of
inorganic fertilizers, annually. By 2020, India has to produce 25 million tones of
banana for exploding population and for export purposes.
For this requirement of inorganic fertilizers is extrapolated to about 25 lakh
tones. The cost of inorganic fertilizers is increasing day by day. Fortunately, N and P
fertilizers are manufactured / mined in India sufficiently based on our needs and their
prices are in our control, but unfortunately, we are depending on foreign countries for
K fertilizers, which account nearly 50 per cent of the cost of inorganic fertilizers
required for banana. There is no chance of reduction in hike in cost of K fertilizers
and hence, in the cost of total fertilizer input for banana. As this is the present
situation, the target, 25 million tones of banana is likely to be out of reach.
On the other hand, banana being a K-loving crop depletes soil K rapidly and
replacement of K in banana soils is not in proportion due to increasing cost of K
fertilizers in the market and hesitation of the banana farmers to application of such a
costly K fertilizers adequately to their soils. Such type of problems leads to sever
nutritional imbalances, which lead to permanent soil health damages.
9
1.3 Soil and factors
1.3.1 Soils<
The fast deterioration of the banana root system takes place when the soil has
one or more of the following characteristics. More than 60% coarse fragments by
volume, high sand content (loamy sand or sand of coarse and very coarse size), very
high clay content without soil structure (massive) or with coarse and very coarse
blocks and prisms. Effective soil depth less than 30 cm is restricted by continuous
rock, massive clay or a shallow permanent water table.
1.3.2 Climatic and topographic factors,
High water table, Frequent water logging of the upper soil horizons (rain and
poor surface drainage), Frequent flooding, Effects in the banana root system.
1.3.3 Effects in the banana root system
The above mentioned factors ends up in weak root system, many dead roots,
few live roots, Short, weak and rotten roots, abundant dead roots, short, shallow and
horizontal roots and few short functional roots with frequent injuries many dead
roots.
1.3.4 Climate and topography
Rainfall is the most important factor involved in banana root system
deterioration.It interacts with topographic factors that may result in severe adverse
conditions for banana root development. The most important of the possible
interactions are flooding, puddles after rains, shallow water tables (permanent or
frequently fluctuating), and areas too close to sea level to be effectively drained.
10
1.3.5 Biological factors
Areas with high nematode populations, other banana root parasitic micro
organisms and insects can cause fast banana root deterioration, especially when the
areas have been previously planted with bananas.
Soil quality is defined as the capability of the soil to function effectively in the
present and future. This integrates physical, chemical and biological soil processes
establishing the most relevant for the production of biomass of sustainable quality
necessary to generate good plant and animal health (Doran and Parkin 1994) [19].
The quantification of the effect of soil in biomass production will depend on
the impact of each individual soil property on the performance of the crop of interest.
The concept applied by Karlen and Stott (1994), to assign weightings to the relevant
soil properties involved in the effects of soil erosion, was applied to evaluate the
effect of these properties in the production of crops other than bananas (Barahona
2000) [9]. The success of the practical application of these concepts (Fernandez
2003[29], Cueva 2003 [15], Orellana 2003)[75] and their usefulness to predict the
performance of several crops has lead to the application of these concepts to banana
cultivation
11
1.4 Research Problem
Plant analysis has been considered as a very promising tool to assess
nutritional requirements of plants for cost effective and environment friendly
agriculture. Diagnosing nutritional status of bananas through plant analysis not only
provides the basis of correct fertilizer requirement of the crop but also guides towards
the nutritional requirements of future crops. The total contents of nutrients in leaves,
and plant parts, compared with Critical Nutrient Range (CNR)5, provide the basis for
interpretation. The Diagnosis and Recommendation Integrated System (DRIS)6 is also
used for interpreting plant analysis data, based on a comparison of calculated
elemental ratio indices with established norms.
The Plant Analysis with Standardized Scores (PASS)7, the most efficient
diagnosis systems, has not been effectively utilized for bananas. The accurate plant
sampling, handling, and analysis of the sample coupled with a thorough knowledge of
cropping history, sampling techniques, soil test data, environmental influences, and
nutrient concentrations favour efficient diagnosis and interpretation system (Menon et
al., 2005) [69]. This, in turn, leads towards more efficient nutrient management and
sustainable crop production. This research based on various critical aspects of the use
of soil variables and plant analysis as a diagnostic tool for banana nutrition
management.
5 Critical Nutrient Range6 Diagnosis and Recommendation Integrated System7 Plant Analysis with Standardized Scores
12
1.5 Aim and Objectives
Banana production systems at the current level of yields are not found to be
sustainable in the long run, as there is significant depletion of plant nutrients in soil.
Build up and maintenance of soil fertility and consequent provision of balanced
nutrition to banana crop is the key to sustain long term banana productivity.
This is the crucial time to encourage judicious application of inorganic
fertilizers and integration of organic nutrient sources based on initial soil test values
and leaf analysis through the development of farmer-friendly fertilizer adjustment
equations and allied computer packages, specific to different banana varieties,
locations, soil series etc. to get a targeted banana yield, based on the financial position
of the farmers, without affecting the soil health and farmers’ wealth adversely. The
following main objectives are given below:-
“ The objective of this research is to build up an ANN based Absolute Update
Technique relating banana yield to soil, weather, and land management factors leaf
nutrients, and to evaluate targeted banana yield based on initial soil test values and
financial position of the farmers using ANN and optimize the quantity of fertilizer
used in the soil based on balanced nutrition concept and sustain soil health by
avoiding inorganic fertilizers in banana cultivation.”
13
1.6 Scope of the Study
Banana production systems at the current level of yields are not found to be
sustainable, in the long run, as there is significant depletion of plant nutrients in soil.
Build up and maintenance of soil fertility and consequent provision of balanced
nutrition to banana crop is the key to sustain long term banana productivity.
In this work a systematic approach has been developed to train different
Artificial Neural Networks with different architectures for banana yield prediction
with new model. To achieve there is a need to develop a more advanced model which
is the ANN Absolute Update Technique. This technique consists of following aspects
Decision-making for agricultural scientists
Advising level of formers
Cost effective Solutions
14
1.7 Limitations of the study
1. Constraints on time and resources restricts to select a cluster of Tamil Nadu soils
and plant nutrients for the study. Hence the results are largely applicable to those
areas where similar conditions prevail.
2. The personal interview method of data collection requires the respondents to recall
from their memories about cultural operations of banana cultivation. Hence, the
findings may be subject to memory lapses of the study.
3. The average price realized during the study year was calculated and used in
converting Production figures from quantities to value terms, although the prices
realized differ from farmer to farmer every year.
4. In the study area, the duration of banana crop yield was different for many farmers.
So the findings of the study permitted to get the same yield.
15
1.8 Significance of the research
Banana is one of the important fruit crops in India. In India, 16.5 million tones
of banana are being produced from 4.5 lakh hectares, with the 16-lakh tones of
inorganic fertilizers, annually. By 2020, India has to produce 25 million tones of
banana for exploding population and for export purposes.
Banana production systems at the current level of yields are not found to be
sustainable, in the long run, as there is major depletion of plant nutrients in soil.
Hence, we are in need of a better and more profitable method of development. In such
a scenario this research gains significance at large. It attempts to set a realistic yield
goal in each part of the field which one of the complicated and critical problems in
precision in the field of agriculture today.
16
1.9 Statement of Problem
In Crop yield history, it is observed that crop production systems are very
complex. Presently, in agriculture, old vegetative methods are used which speaks on
the analysis of past data only. It does not have any relevancy for future prediction
with enough confidence. In this scenario, the researcher has identified the need of a
better and more profitable method of development which is discussed in this research
titled as “SOIL TEST BASED INTEGRATED NUTRIENT TAILORING FOR
OPTIMUM BANANA PRODUCTION AND SUSTAINABLE SOIL HEALTH
USING ARTIFICAL NEURAL NETWORKS”.
17
Structure of Thesis
The thesis is split up into several chapters as follows:
Chapter 1 Introduction
Chapter 2 Review of literature
Chapter 3 Survey Analysis and Design
Chapter 4 Methodology
Chapter 5 Result and Discussion
Chapter 6 Comprehensive conclusion and Scope of the future work
Chapter 7 Reference
18
2. REVIEW OF LITERATUREA review of the research work done in the past relating to the present study
has been presented in this chapter. Number of studies conducted in banana yield. A
long with a review of literature is presented under the following sub titles.
2.1 Soil based banana plant yield
2.2 Artificial Neural Networks based on other crops yield methods
2.3 Applications of Fertility Gradient approach for various crop yield prediction
2.4 Costs and returns of tissue culture banana and sucker propagated banana.
2.5 Resource use efficiency in tissue culture banana and sucker propagated
banana.
2.6 Marketing channels and marketing costs.
2.7 Problems in production and Marketing of banana.
2.1 Soil based banana plant yield
Delvaux (1995) [16] suggested that soil fertility (health), was a poorly defined
concept that not only relied on soil chemical, physical and biological properties, and
their interaction with the plant community, but on management practices, farming
skills and economics.
Doran and Parkin (1996) [19] defined soil health as “the capacity of a soil to
function within an ecosystem and land use boundary, to sustain biological
productivity, maintain environmental quality and promote plant and animal health”.
Van Bruggen and Semenov (2000) suggested that a healthy soil is a stable soil
with resilience to stress, high biological diversity and internal cycling of high
19
amounts of nutrients. Knowledge of the function of the soil ecosystem is a basic
requirement for soil stewardship (Ferris et al. 2001) [30].
Nematodes are components of the soil ecosystem that interacts with biotic
and a biotic soil factors (Yeates 1979). Because of this interaction, nematodes are
excellent bio-indicators of soil health, because they form a dominant group of
organisms in all soil types, have high abundance, high biodiversity and play an
important role in recycling within the soil (Neher 2001, Schloter et al. 2003 [82]).
Nematodes are heterotrophy, higher in the food chain than micro-organisms and so
serve as integrators of soil properties related to their food source, predators and
parasites (Ferris et al. 2001 [30], Neher 2001). Nematode diversity tends to be the
greatest in ecosystems with the least disturbance (Yeates 1999) [95].
The disturbance to the soil by environmental or land management practices
changes the composition of nematodes (Bongers 1990 [12], Yeates and Bongers
1999, Ferris et al. 2001) [30]. There are a number of indices derived from nematode
community analysis that can be used to determine the impact of management changes
on the soil ecosystem (Bongers 1990, Yeates and Bongers 1999, Ferris et al. 2001)
[30].
However, the finest use of nematodes is that they serve as the indicators of
soil ecosystem. Health and banana management is not a practical tool for farmers, as
it requires specialized knowledge and equipment (Neher 2001). Doran (2002) [21]
suggested linking “science to practice” in assessing the sustainability of land
management practices, by the use of simple indicators of soil quality and health that
have meaning for farmers. To embrace changes in environmental management of
their land, farmers need to understand why they need to change (Marsh 1998) [66].
The best way to achieve this is by the use of participatory research strategies using
simple on-farm techniques (Freebairn and King 2003[34], Lobry de Bruyn and Abbey
2003). A basic set of soil quality indicators was developed by J.W. Doran (USDA-
ARS, Lincoln, NE), and developed into an on farm test kit
“http://soils.usda.gov/sqi/soil_quality/assessment/kit2.
20
html).The basic set of soil parameters has been used to measure the effects of changes
in soil management on agricultural crops (Sarrantonio et al. 1996[81], Stamatiadis et
al. 1999) but not on bananas.
One important soil characteristic that is not easily measured is soil organic
carbon. Widmer et al. (2002) [98] suggested that maintenance of high concentrations
of organic matter,especially the active fraction, greatly improves the physical,
chemical and biological properties of soils leading to increased productivity. Tropical
soils used in banana production tend to have high soil water contents and high soil
temperatures, which are favourable for organic matter decomposition (Sikora and
Stott 1996) [85]. Additionally, intensive cultivation of the soil in preparation for
planting bananas in north Queensland may also be reducing soil carbon. A simple on
farm test to determine soil organic carbon is needed to allow monitoring and linking
in to other soil health indicators.
Plant analysis is considered as a very practical approach for diagnosing
nutritional disorders and formulating fertilizer recommendations (Kelling et al.,
2000[49]; Self, 2005). Plant analysis, in conjunction with soil testing, becomes a
highly useful tool not only in diagnosing the nutritional status but also an aid in
management decisions for improving the crop nutrition (Rashid, 2005). Plant analysis
is the quantitative analysis of the total nutrient content in a plant tissue, based on the
principle that the amount of a nutrient in diagnostic plant parts indicates the soil’s
ability to supply that nutrient and is directly related to the available nutrient status in
the soil (Malavolta, 1994[62];Kelling et al., 2000[49]; Havlin et al., 2004[37];
Rashid, 2005[80]). It is a very practical and useful technique for fruit trees and long
duration crops (Rashid, 2005). Hence, it seems quite convenient and appealing for
bananas also.
Bananas are heavy feeder of nutrients (Jones, 1998)[47] and thus need
balanced nutrition for optimum growth and fruit production, and in turn potential
yields. A deficiency or excess of nutrients can cause substantial damage to the plant
(Memon et al., 2001)[68]. The early (until the mid-1960s) researches on banana
21
nutrition had concentrated on the description of symptoms of nutrient imbalance and
the conduct of field experiments comparing response to rates of applied fertilizer on
a range of soil types. During last three decades, scientists attempted to understand
more clearly the role of nutrients in the growth and development of bananas. Field
studies of fertilizer response are still being conducted, but attempts to relate nutrient
concentrations in the soil and plant to yield have complemented this work. Analysis
of plant parts for mineral elements and the attempt to set standards for interpreting
leaf analysis data came to the fore in the late 1960s and early 1970s. However, each
researcher approached the problem differently, probably reflecting a lack of unifying
concepts in the understanding of the growth and nutrition of bananas, until Martin-
Prevel (1974, 1977) [64][65] initiated the formation of an International Group on
Mineral Nutrition of the Banana that resulted in a suggested International Reference
Method for sampling in banana fertilizer experiments. In this paper, the important
aspects of banana nutrition management through plant analysis have been reviewed.
Sampling procedures have been investigated by many researchers (Dumas,
1959 [25]; Twyford & Coulter, 1964; Martin-Prevel et al., 1969; Lahav, 1970[53];
Turner & Barkus, 1977). Earlier, researchers at the Jamaica Banana Board (Hewitt,
1953; Hewitt & Osborne, 1962) and IRFA, Guinea (Dumas & Martin-Prevel, 1958;
Dumas, 1960a[26]), used different approaches and defined some of the problems
associated with sampling in banana. It was thus difficult to perceive indisputable
overall advantage in either one method or the other and hence many workers
preferred to establish a procedure well suited to their own special circumstances. In
two decades, a variety of procedures were used. Later on, Martin-Prevel (1977) [65]
came up with a measure of uniformity to sampling methods by surveying the methods
used in different countries.
2.2 Artificial Neural Networks based on other crops yield methodsDrummond et al.,(1995)[23] compared several methods for predicting crop
yield based on soil properties. They noted that the process of understanding yield
variability is made extremely difficult by the number of factors that affect yield. They
22
used several multiple linear regression methods, such as multiple linear regression
(MLR), R 2 = 0.42; stepwise MLR (SMLR), R 2 = 0.43; partial least squares
regression (PLSR), R 2 = 0.43; projection pursuit regression (PPR), R 2 = 0.73; and
back-propagation neural network (BPN), R 2 = 0.67 for modeling the relationship
between corn yield or soybean yield and soil properties. They concluded that less-
complex statistical methods, such as standard correlation, did not seem to be
particularly useful in understanding yield variability. The correlation matrices
described each factor’s linear relationship to yield. However, when complex
nonlinear relationships between factors exist, correlation might provide inaccurate
and even misleading information about these relationships.
Dudley Smith et al.,(1984) compared with Marrow plots yield prediction for
soybean-corn rotation. The 32-hectare field includes five soil types: Herrick silt loam,
Virden silt loam, Virden silty clay loam, Oconee silt loam, and Harrison silt loam.
The field was divided into 1041 grid points with 18.3 m 18.3 m spacing Compared
with the Morrow Plots, the Dudley Smith farm had much greater spatial distribution
but only two years of temporal data were available. When used on the Dudley-Smith
farm without retraining. The ANN gave an RMS yield prediction error of 41.3%, i.e.,
the prediction accuracy fell to 58.7%. Through the ANN re-training, the prediction
accuracy was increased to 83% for the Dudley-Smith farm field.
ANN is an alternative for building a knowledge discovery environment for a
crop production system. An ANN can use yield history with measured input factors
for automatic learning and automatic generation of a system model. In the past few
years, several yield simulation models have been built. Ambuel et al., (1994) [3] used
23
a fuzzy logic expert system to predict corn yields with promising results. The
functional relationship using the fuzzy logic expert system was expressed
linguistically instead of mathematically. The authors suggested the use of a neural
network to predict within field yields.
Crop yield history suggests that crop production systems are very complex.
Both process-oriented crop growth models and traditional statistical methods can be
used to study crop growth and yield response to environment and management. For
example, Paz et al. (1999, 1998) [76] developed a technique to characterize corn yield
variability using the CERES-Maize process-oriented crop growth model, and to
characterize soybean yield variability using the CROPGRO-Soybean process-oriented
crop growth model.
The back-propagation neural network is a universal approximator (Haykin,
1994) [38]. Given sufficient hidden units, multi-layer feed-forward network
architectures can approximate virtually any function of interest to any desired degree
of accuracy (White et al., 1992) [97].
2.3 Applications of Fertility Gradient approach for various crop
yield prediction
B.S KADAM AND K.R. SONAR et al (2006) were conducting a research on
post-monsoon Onion using fertility gradient approach (Ramamoorthy et al. 1967) on
otur soil series. It has 21 selected treatment combinations out of 5 levels of nutrients
N, 4 levels of nutrients P, 3 levels of K with six control plots. Efficiency of soil
nutrient was 11.25, 55.35 and 7.37% of N, P, K while that of fertilizer N, P, K were
21.01, 29.35 and 66.18 respectively. Ramamoorthy et al.,(1967 ) reported that limit
24
of fertilizer application was dependent on these parameters. It was on this basis that
prescription method of fertilizer recommendation for targeted yields of corn was
advocated by Truog(1960). Fertilizer rates increased with increasing yield targets on
onion and fertilizer rates decreased with increasing the soil test values. Thus in the
targeted yield concept potential and soil test values were taken into account while
making fertilizer recommendations. The result of the two follow-up trails on onion
otur and sawargan series showed that yield targets of 30, 40 and 50 t ha-1 were
achieved. The highest yield 53.5 t ha-1 and profit rs.90300 were observed under 50 t
ha-1 yield target on onion followed by 40 t ha-1 targeted yield approach.
An increase in profits over control was observed with increasing yield targets
from 30-50 t ha-1 which might be due to efficiency factors tended to increase with
increase in crop yields (Sekhon et al., 1977). Similar results were also reported by
Hariprakasha Roa and Subramanian (1994) and Anonymous (2000) in vegetable
crops. From These studies it is possible to make fertilizer recommendations for onion
yield prediction to the formers considering their financial conditions.
2.4 Costs and returns of tissue culture banana and sucker
propagated bananaSenthilathan and Srinivasan (1994)[83], estimated the cost and returns of
Poovan cultivar banana production in Thrichirapalli district of Tamil Nadu, over a
period of three years. Total cost of cultivation per hectare was Rs. 1, 24,668.11, with
the gross income of Rs. 2,86,913.80 and there by the net income worked out to be Rs.
1, 62,235.69 per hectare. The study clearly showed the high profitability of varity
Poovan banana with a high benefit costratio 2.3: 1 in the study area.
Maurya et al. (1996) [67] studied the profitability of banana production in
Hajipur district of Bihar state, during 1993-94 based on sample of sixty banana
25
growers selected from five villages in the district. The study revealed that, banana
production was the most profitable crop production activity in the study area, as it
provided a net income of Rs. 29,748.05 per hectare with a total expenditure of Rs.
20,160.70 and gross income of Rs. 49,958.75.
Dhakate (1996) [18]studied the economics of banana production in Akola
district of Maharashtra. The data was collected from 75 banana growing farmers
through personal interview method. The average output per hectare was 40.29 tonnes
valued at Rs.71,743.32 per hectare, gross returns ranged from Rs. 69,894.78 on large
farms to Rs. 74,521.59 on small farms. Per hectare profit at cost ‘C’ for the sample as
a whole was Rs.
19,533.79 and it ranged from Rs. 17,685.20 in large size group to Rs. 18, 557.48 in
small size group.
Sudarshan (1998) [88] in the project conducted on an experimental farm in
Bangalore reported that tissue culture banana had a world record of 6,900 plantlets
per hectare. The tissue culture banana plantlets give very yields compared to sucker
based plants of the same variety. Compared to average national yields per plant of 9
to 10 Kg (bunch weight) and average commercial banana produces yield per plant of
15 to 20 Kg in sucker based crop, the tissue cultured plantlets yield a bunch weight of
40 to 60 Kg per plant. The plantlets yield 175 tonnes as against 45 tonnes of
conventional commercial sucker based banana horticulture in India. The estimated
revenue per crop of 11 months was Rs. 12.5 lakhs per hectare could be obtained at a
conservative price of Rs. 5 per Kg of banana. The revenues are further augmented by
selling stem cores, which may fetch Rs. 3 to 5 per Kg at whole sale. The tissue culture
daughter suckers can also be sold, which may fetch a price of Rs. 5 per sucker.
More (1999) [72] studied the economics of production marketing of banana in
Marathawada region of Maharashtra state, he found that the cost of cultivation of
banana per hectare was higher on small farms (Rs. 32,294.72) compared to large
farms (Rs. 76,610.06) due to inefficiency in utilization of bullock labour, machine
labour, human labour and manure and fertilizer in case of large farmers. The gross
26
income per hectare was higher in large farmers (Rs. 1, 42,885.30) compared to small
farmers (Rs. 1, 40,696.80) due to higher yields in large farmers.
Qaim (1999) [78 ] studied Socio-Economic impact of tissue culture
technology in banana production in Kenya. The study revealed that, the cost of
production of tissue culture banana was significantly higher (Increase in cost was
130% in small scale, 118% in medium scale and 92% in large scale growers)
compared to banana production without tissue culture. This was due to higher labor
intensity besides the use of more inputs. Accordingly Yields and incomes obtained
per hectare of banana were also higher (increase in yield was 150% in small scale,
132% in medium scale and 93% in large scale growers. Increase in income was 156%
in small scale, 145% in medium scale growers and 106% in large sale growers) on
these farms. The results also revealed that adoption of tissue culture could bring about
substantial increase in yield for all the three types of farmers (small, medium and
large). In relative terms the potential gains are most pronounced for small farmers.
Kameswara Rao (2000) [48]compared economics of banana and sugarcane
cultivation in Tungabhadra command area of Karnataka. The results revealed that, the
per hectare total cost was higher in banana crop (Rs. 71,513.04) than in sugarcane
crop (Rs. 65,496.12). The per hectare gross income and net income generated in
banana cultivation were also higher (Rs. 1, 13,377.57 and Rs. 41, 864.53
respectively) as compared to sugarcane crop (Rs. 81,382.74 and Rs. 15,886.62)
respectively. The benefit cost ratio at cost ‘D’ was lower in sugarcane (1.24)
compared to banana cultivation (1.59).
Mishra et al (2000) [71] conducted a study on production and marketing of
banana in Gorakhpur district of Uttar Pradesh. The researchers estimated the total per
hectare cost of production of banana on small, medium and large farmers at Rs.
36,281.50, Rs. 37,820.50 and Rs. 38,447.50 respectively, with average cost of Rs.
37,516.50 per hectare. The per hectare average gross returns were Rs. 71,133.33,
which was higher on large farms (Rs.73,400) followed by medium farms (Rs. 72,250)
and small farms (Rs. 67,750). The average input output ratio was 1: 1.89. Anonymous
27
(2000) [5], the tissue culture banana crop cycle comprised of three crops in two years.
The cost of cultivation of main crop per hectare was about Rs. 235,000 and second
and third crops were Rs. 100,000 each. The expected return out of three crops was
about Rs. 15.0 lakhs per hectare and average returns per hectare of banana was Rs.
5.0 lakhs and Rs. 7.5.0 lakhs per year.
Stephen et al (2002) [ 87] compared the socio-economic impact of tissue
culture banana with non tissue culture banana in Kenya. They found that, tissue
culture banana production was relatively more capital intensive than sucker
propagated banana production. However, tissue culture banana production was found
to be more profitable (yield from sucker propagated banana production was only 60%
of that of yield from tissue culture banana production) compared to non sucker
banana production. Shivanad (2002) [84] in his study on performance of banana
plantations in northern Karnataka, revealed that the establishment cost of banana
plantations was Rs. 74, 759 per hectare, of which 50 per cent was spent on suckers
and stalking. The cultivation of sucker propagated banana was found profitable in
northern Karnataka with a net profit of Rs. 85,266 per hectare per year.
Guledgudda et al., (2002) [36] conducted study on economics of banana
cultivation and its marketing in Haveri district of Karnataka, reported that the variable
cost incurred by producer was Rs. 54,502.81 per hectare which was accounted to 65
per cent of total cost. Among variable costs, the human labour was found to be the
major item of cost, which accounted for 18 per cent. On an average farmers got 175
quintals of banana yield as main product valued at Rs. 1, 54,375 and farmers have
realized Rs. 30,000 by selling suckers, the gross returns from banana cultivation were
Rs. 1, 84,375 per hectare. The net returns realized by farmers were Rs. 1, 00,545.96
with a B: C ratio of 2.19.
Mali et al (2003) [63], in their study on economics of production and
marketing of banana in Jalgaon district of Maharashtra. The worked out cost per
hectare of banana cultivation was Rs. 1, 33,477.36, the gross returns per hectare of
banana at Rs. 2, 14,867.24 and net returns at Rs. 6, 67, 61.87. Florence Wambugu
(2004) [31] , in the study, compared tissue culture and conventional banana. The
28
study revealed that, the average establishment cost per farm (0.2 hectares) was US$
200 in conventional banana and US$ 600 in tissue culture banana, average annual
yield per farm was 5 tonnes in conventional banana and 10 tonnes in tissue culture
banana. The average annual net profit per farm was US $ 600 in conventional banana
and US$ 1800 in tissue culture banana. This means that there were more benefits of
adopting the tissue culture technology compared with staying with the conventional
bananas.
Silva et al (2005) [86] carried out a study in Brazil, to survey the potential of
banana cv. Apple cultivation in the region, as well as to determine the technical and
economical indicators of two production systems, both using micro propagated and
conventional seedlings. The results of economic analysis turned out to be quite
satisfactory in this region for both production systems however the net income
obtained from the utilization of micro propagated seedlings was 34 per cent higher
than the one obtained from the conventional system.
Alagumani (2005) [2] in the study on economic analysis of tissue- cultured
banana and sucker-propagated banana in Theni district of Tamil Nadu, revealed that,
per hectare cost was high in case of tissue culture banana (Rs. 1, 41, 040) compared
to sucker propagated banana (Rs. 1, 08, 294). The net income was also high in case of
tissue culture banana (Rs. 1, 12, 262) compared to sucker propagated banana (Rs. 78,
855), clearly indicating the higher profitability of tissue culture banana production
compared to sucker propagated banana production.
Rane and Bagade (2006) [79 ] studied economics of production and marketing
of banana in Sindhudurg district of Maharashtra. The study revealed that the per
hectare cost at cost C in Dodamarg and Sawantadi tahsil were Rs. 1.52 lakhs and Rs.
1.53 lakh respectively. In Dodamarg tahsil banana was grown as a sole crop where
per hectare cost of cultivation was Rs. 1.28 lakh and in Sawantadi tahsil the per
hectare cost was Rs. 1.15 lakh. The benefit cost ratio in Dodamarg tahsil and
Sawantadi tahsil were 2.20 and 2.33 respectively. The average benefit cost ratio of
banana cultivation was 2.27.
29
2.5 Resource use efficiency in banana production
Venkatesha Reddy (1982)[91]employed Cobb-Douglas type of production
function to examine the resource productivities and efficiency for planted and ratoon
crops of Robusta and dwarf Cavendish varieties of banana separately. The gross
return was the dependent variable while land (hectare), Plant population (number),
labours (man days), manures and fertilizers (Rupees) and plant protection chemicals
(Rupees) were independent variables. The marketing channels were treated as dummy
variables. The analysis indicated that 95 per cent variation in gross return was
explained by above included independent variables.
Thomas and Gupta (1987) [90] studied the economics of banana cultivation in
Kottayam district of Kerala. The Cobb-Douglas production function was used and the
results showed increasing returns to scale. The response of gross income to an
increase in the expenditure on suckers, plant protection chemicals, propping
materials, baskets, transportation and marketing was highly significant and positive,
and the marginal value product of human labour, and manures and fertilizers were
lesser than marginal input prices. The study revealed that the increase in suckers,
plant protection chemicals, propping materials would generate a substantial level of
additional income with a negative amount of additional expenditure.
Dhakate (1996) [18 ] , studied the resource use efficiency and functional
analysis by using modified Cobb-Douglas type of production function. Bullock and
machinery labour , irrigation and maintenance charges , suckers and human labour
charges were considered as independent variables and that of yield as dependent
variable. The results thus indicated that the six variables included in the function
explained about 15 per cent of variation in the output at overall level. In small,
medium and large size groups, it was explained about 69, 29 and 29 per cent of
variation in output respectively. It was observed that in small size group manure and
fertilizers influenced the yields significantly, while at overall level, human labour
influenced the yields significantly.
30
More (1999) [72] studied the economics of production and marketing of
banana in Maharashtra state. Cobb-douglas type of production function was used to
determine the level of resource use efficiency for the banana crops of small, large and
pooled farmers. The independent variables included in the function were land, human
labour, machine labour, farm yard manure, nitrogen, potash, capital, irrigation and
bullock labour. The dependent variable was yield of banana. The coefficients of
multiple determinations were 73, 67 and 85 per cent respectively for all the three
categories of farmers. Land and capital were significantly influenced on yield in all
the three categories of farmers and others were non-significant. The MVP to MFC
ratios for land, phosphorus, capital and bullock labour in all the categories and human
labour machine labour in large farmers category were more than unity, indicating that
under-utilization of these resources.
Kameswara Rao (2000) [48], studied resource productivity of banana crop in
Tungabhadra command area of Karnataka. Researcher employed Cobb-Douglas type
of production function in which banana yield (tonnes) as dependent variable and Land
(hectares), human labour (man days), bullock labour (pair days), suckers (tonnes),
irrigation (numbers), and value of manures and fertilizers (Rs. ) as independent
variables. The results revealed that the land was underutilized to the extent of 25.35
per cent. The regression coefficient of human labour for small labour was significant
at 1 per cent level and the same was non significant for large and pooled farmers. The
regression coefficient of irrigation for small and large farmers was significant at 5 per
cent level. The regression coefficient for land, bullock labour, sucker and manures
and fertilizers were non significant in all categories of farmers. The variation in
banana production was explained by selected independent variables in small and large
farmers group.
Shivanand (2002) [84] studied the resource use efficiency of banana crop in
northern Karnataka. He employed Cobb-Douglas type of production function, where
banana yield as dependent and land, labour, FYM, bullock labour, fertilizers plant
protection chemicals as independent variables. The study showed that land, labour
31
and plant protection chemicals have significantly influenced the production of banana
as indicated by their significant regression coefficients of 0.672, 0.472, and 0.172
respectively in the study area. The MVP to MFC ratio were positive and more than
one for land (7.890), labour (5.321) and FYM (1.34) , where as it was less than one in
case of fertilizers (0.871), bullock labour (-401.94) and plant protection chemicals (-
2.73).
Yadav et al., (2004) [93] made comparative study of resource use efficiencies
and resource productivities of traditional and tissue culture banana production in
Maharashtra state. The regression co-efficient of area, FYM and potash were positive
and significant at 10 per cent level, indicating, scope to increase level of those inputs
to step up the productivity. The sum of elasticities of production was equal to unity;
reveal that constant returns to scale in traditional method of banana production. In
tissue culture banana, the functional analysis revealed that, the regression co-efficient
of plantlet was highly significant, there by indicating scope to increase the level of
plantlet. The sum of elasticities of production was equal to unity showing constant
returns to scale. MVP/ MFC ratio for inputs namely sucker nitrogen and bullock
labour was greater than unity referring that efficient use of these resources.
Alagumani (2005) [2] found that, in tissue-culture banana, the co-efficient of
Plantlets, manures, and fertilizer were positive and significant at 1 per cent level.
Labour cost had negative and non-significant influence on gross income. Sum of
elasticities of resources shown that constant returns to scale. Where as in sucker-
propagated banana the co-efficient of sucker and fertilizer costs were positive and
significant at one per cent level. The sum of elasticities of resources shown that
decreasing returns to scale.
More et al (2005) [73 ]studied on Labour utilization and input use pattern in
banana cultivation in Maharashtra. Data were collected from 120 banana growers in
Nanded and Parbhani districts determine the labour utilization and input use pattern
(fertilizer and irrigation use) in banana cultivation. The study revealed that the major
proportion of human labour was used for irrigating the banana crop. Hence, there is a
32
need to encourage farmers to adopt the drip irrigation method, which is somewhat
costly but labour-saving.
2.6 Marketing channels and marketing costsThe marketing channels and marketing costs of banana produced by both the
methods are dealt in similar way, because there is no difference found in the quality
of output.
Senthilnathan and Srinivasan (1994) [83 ]identified the channels of banana
marketing in Trichirapalli district of Tamil Nadu. Identified channels were Channel-I:
Farmer _ pre harvest contractor _Secondary wholesaler _ retailer _consumer
Channel-II: Farmer _ pre harvest contractor _ commission agent _ wholesale _retailer
_ consumer.Channel-III: Regulated market wholesaler_ retailer _ consumer.Channel-
IV: Farmer _ Regulated market secondary wholesaler_ retailer _ consumer.The study
found that, channel-III was best among four channels.
More (1999) [72] in his study on economics of production and marketing of
banana in Maharashtra state, researcher identified two major marketing channels in
the study area through which bananas move from producer to consumer. They were
Channel-II: Producer _ commission agent cum wholesale _ retailer _
consumer.Channel-I: Producer _ commission agent _ distant markets.The marketing
cost incurred by producer seller was Rs. 15.17 per quintal of banana.
Kameswara Rao (2000)[ 48] in the study on comparative economics of banana
and sugarcane cultivation in Tungabhadra command area of Karnataka, identified two
marketing channels in the study area, namely Channel-I: Producer_ commission
agents_ Wholesaler_ Retailer_ Consumer Channel-II: Producer_ Village level trader_
Wholesaler_ Retailer_ Consumer. In Channel-I, the total marketing cost incurred by
producer-seller was Rs. 23.44 per quintal of banana. In Channel-II, the producer-
seller has not borne any marketing cost.
33
Mishra et al (2000)[71] study on production and marketing of banana in
Gorakhpur district of Uttar Pradesh. The researchers identified that the small farmers
were selling their produce to wholesalers (20% of produce), village trader (20%), pre
harvest contractor (25%), direct sale local market (20%) and commission agent-cum –
wholesaler (15%). Medium farmers sold their produce to the wholesalers (15%),
village traders (20%), pre harvest contractors (25%), direct sale in local market (25%)
and to the commission agents (20%). And large farmers sold their produce to
wholesalers (10%), village trader (15%), Pre harvest contractor (25%), direct sale in
local market (30%) and commission agent cum wholesalers (25%). The marketing
cost per quintal of banana produce incurred by the producer in fourth channel was Rs.
37.50 (9.87% of total marketing cost) and Rs. 24.25 (6.14%).
Guledgudda et al., (2002) [36]conducted study on economics of banana
cultivation and its marketing in Haveri district of Karnataka. The results of the study
revealed that farmers in the study area followed three distinct marketing channels to
sell their bananas. Those channels were Channel-I: Farmer _ pre harvest contractor’s
_ commission agent _ wholesale _retailer _ consumer.Channel-II: Farmer _
commission agent _ wholesale _ retailer _ consumer Channel-III: Farmer _ retailer _
consumer. Farmers have spent Rs. 1.50 per bunch of banana marketing in channel-I,
Rs. 2.50 per bunch in channel- II and Rs. 10.25 in case of channel-III.
Shivanand (2002) [84] in the study on performance of banana plantations in
Northern Karnataka- An economic analysis, identified two major marketing channels
namely, Channel-I: Producer_ commission agent cum Wholesaler_ Retailer_
Consumer.Channel-II: Producer_ Village level trader_ Commission agent cum
Wholesaler _ Retailer_ Consumer. Among the two channels identified channel-I was
found predominant over channel-II in marketing of banana in study area. On an
average producer has incurred a marketing cost of Rs. 9.50 in channel-I farmers has
not incurred any marketing costs in channel-II.
34
Stephen.et.al., (2002) [87] studied the Socio-economic impact of tissue
culture banana compared with non tissue culture banana in Kenya. It showed that the
farmers primarily sell their bananas in the form to traders and other marketing
intermediaries (popularly referred as BROKERS), either at farm gate or at their local
trading centers. Many of the produce buers (brokers) come from major urban market
centers. Depending on the distance from farm to nearest trading center, farmers pay
between KShs.10 and KShs.30 per bunch of bananas plus between KShs.20 and
KShs.50 each way as consumer fare and KShs.10 per bunch as Cess by the local
authority for selling at local trading centers. Hence it will costs between KShs.50 and
KShs.100 to deliver and sell a bunch of bananas at local trading centers.
Gajanana (2002) [35] studied Marketing practices and post-harvest loss
assessment of banana var Poovan in Tamil Nadu. The study revealed the marketing
practices of Poovan variety of banana in Trichy district of Tamil Nadu. Data on
marketing practices were collected during March-April 2001 from the growers of
Trichy and Lalgudi taluks of the study district. It can be inferred from the whole
analysis that the farmers use value judgement by resorting to field sale to the agents
of the distant market (Bangalore regulated market) wholesaler for getting higher
price. A need is suggested to convert all unregulated markets like Trichy into
regulated markets for the benefit of farmers throughout the country.
Mali et al., (2003) [63], studied economics of production and marketing of
banana in Jalgaon district of Maharashtra. The marketing of banana in Jalgaon district
is done through three marketing channels viz., Delhi market through co-operative
marketing societies, private traders through co-operative fruit sale societies and local
merchants/group sale agencies. It was identified that the average per quintal
marketing cost was Rs. 27.55. It includes wages paid to labour for transportation from
field to bund/road commission charges of market agencies and transportation charges
of the produce from field to Railway station.
35
Vinod-Wanjari and Ladaniya (2004)[92], examined marketing costs, margins
and important marketing channels for bananas, cv. 'Basrai' (Dwarf cavendish), on the
basis of data collected from growers, cooperative societies and intermediaries in
selected districts of India (Jalgaon and Nagpur districts in Maharashtra, and
Burhanpur district, Madhya Pradesh). Farmers sell their standing crop to pre-harvest
contractors and also to cooperative societies and commission agents. The net price
received by the farmer is slightly less when produce is sold through a cooperative
society since the society charges a higher commission than private commission
agents. Transportation cost increased with distance between production area and
market and this increased the marketing.
Ajay Verma and Singh (2004) survey identified common marketing channels
in different states of the country. In Pune and hazipur: Producer _
wholesaler/commission agent _Retailer _ConsumerProducer _ trader_ trade
wholesaler _Retailer _ Consumer.In Ranchi: Producer _ distant market. Producer _
Retailer _ Consumer Producer _ wholesaler/commission agent _ Retailer _ Consumer
In Guawahati: Producer _ contract _ trader _ distant market. Producer _ contractor _
trader _ wholesalers _ Retailers. Producer _ wholesaler or commission agent_ Agent
retailer. Bhubaneswar: Producer _ trader _ distant market. Producer _ trader _
wholesaler _ Retailer.
36
2.7 Problems in production and marketing of banana
Senthilnathan, and Srinivasan (1994)[ 83], estimated the cost and returns of
Poovan cultivar banana production in Thrichirapalli district of Tamil Nadu. The study
revealed that, in Trichy taluk twenty per of cent farmers expressed high initial
investment, sixteen percent farmers expressed problem of heavy wind damage
similarly twelve price fluctuations and ten disease problems. In Lalgudi taluk there
were seventeen high initial investment, eleven price fluctuation, thirteen disease
incidence and nine wind damage. In Kulikathi taluk two disease incidence, eighteen
wind damage and fourteen price fluctuations.
Qaim (1999) [78] studied Socio-Economic impact of Tissue Culture (TC)
technology in banana production in Kenya. The study revealed that, due to high
expenses for the technology itself and for complementary inputs, small farms were
facing the most severe adoption constraints.
More (1999) [ 72] studied the economics of production marketing of banana
in Marathawada region of Maharashtra state. The study identified problems faced by
the farmers that, all the farmers in the study area were facing the problem of Musa
sercospora disease. The other major problems were high labour wages, non
availability of quality planting materials at right time at reasonable price and non
availability of adequate technical assistance from experts on behalf of government.
The problems in marketing were spatial variation in the prices creating uncertainity
among cultivatiors in choosing the markets for sale of produce. The higher
transportation cost was also one of the major marketing problems in marketing of
banana in the study area. Inadequate availability of the loan at right time by the
financial institutions was the main problem in the production of banana in the study
area.
37
Mishra et al., (2000) [71] in their study on production and marketing of
banana in Gorakhpur district of Uttar Pradesh, identified problems faced by the
farmers in the production and marketing of banana, unavailability of quality suckers
and high cost of seed suckers, high cost of transportation, lower ruling price for
produce due to unavailability of adequate storage facilities and weak finance
structure. The problem of poor supply of power electric power in critical period,
unavailability of fertilizers and insecticides at reasonable prices.
Kameswara Rao (2000) [ 48]studied the problems of production and
marketing of banana in Tungabhadra command area. The study revealed that, the
major problems faced by the 85 per cent of the farmers was non availability of
sufficient irrigation water. 73 per cent of farmers were opined that higher prices of
fertilizers, 68 per cent of the farmers were facing the problem of non availability of
quality planting material. The other major problems in production of banana in study
area were labour shortage in peak time, hazards of soil salinity, hail storms of heavy
winds. The major financing problems in the study area were available loan was
inadequate, high procedural complication of loan and high rate of interest. The major
problems in marketing of banana in study area were high price fluctuations, high
transportation cost, delayed payments on sale proceeds by the trader/businessman and
high commission of intermediaries.
Begum and Raha (2002)[ 10], studied on Marketing of banana in selected
areas of Bangladesh. The existing marketing system for bananas in selected areas of
Bogra district,Bangladesh, was examined, based on data from 40 market
intermediaries. Also examined were the marketing costs and margins at different
levels of banana marketing and the existing marketing constraints. Results revealed
that banana marketing is a profitable venture and major marketing problems are price
instability, lack of capital, inadequate facilities, and lack of adequate market
information.
38
Guledgudda et al., (2002) [36] conducted a study on economics of banana
cultivation and its marketing in Haveri district of Karnataka. The study identified
production problems like lack of technical know-how, scarcity of labour, pest and
diseases, lack of adequate credit facility, and scarcity of water. The farmers in the
study area expressed also marketing problems like involvement of intermediaries,
lack of storage facilities and inadequate transportation.
Stephen et al (2002) [87] studied the Socio-economic impact of tissue culture
banana compared with non tissue culture banana in Kenya. The study revealed that
the tissue culture banana producers appear to be constrained by capital for investment
in irrigation facilities and acquisition of fertilizers or organic manures to produce
good banana crop. Lack of organized marketing facilities makes exploitation of
banana producers by traders/brokers fairly easily.
Shivanad (2002) [ 84]studied the performance of banana plantations in
northern Karnataka. The study revealed as perceived by the farmers the major
problems in cultivation of banana were severe incidence of Musa sercospora disease
in all the districts of northern Karnataka, the disease lead to heavy crop losses. Erratic
onset of monsoon was another problem in Belgaum district affecting banana
plantations. In Gulbarga district the non availability of labour and high labour wages
and non availability of technical assistance for improved cultivation of banana
possesses severe problem in production of banana. In marketing of banana farmers
were facing delayed payments of sale proceeds, high cost of transportation of
produce, wide price fluctuations and high commission charges as major problems.
,
Mali et al., (2003)[ 63], studied economics of production and marketing of
banana in Jalgaon district of Maharashtra. The study identified that high cost of
transportation, non availability of sufficient credit by the institutions in time, high
price fluctuations, the problem of cheating in weighing of produce and lack of
suitable grading of the produce according to quality as main problems in production
and marketing.
39
Alagumani (2005) [2] in study on economic analysis of tissue- cultured
banana and sucker-propagated banana, in Theni district of Tamil Nadu. The study
revealed that, the risk in cultivation of banana using tissue culture plantlets was lower
than that of sucker propagated banana production. The constraints in tissue culture
banana production were cost of tissue culture plantlets were very higher, and few
farmers were also expressed problem of marketing of big size bunches obtained from
tissue culture banana.
Rane and Bagade (2006) [79] studied economics of production and marketing
of banana in Sindhudurg district of Maharashtra, the study reveals that farmers were
facing the problem of disease i.e. bunchy top disease of banana and also farmers were
facing the problem of pest i.e. aphids of banana in production of banana.
40
2.2 Synthesis of the Literature Review
This analysis proceeded towards the literature survey for the research work,
started working on Banana production systems at the current level of yields are not
found to be sustainable, in the long run, as there is significant reduction of plant
nutrients in soil. An Artificial Neural Network was used to build a crop yield
prediction model is Absolute Update Technique. This Technique involved for various
analyses and reduces the more number of iterations and used to drastically reduce the
dimension of the network and computational effort and farmer-friendly and based on
the financial position of the farmers, without affecting the soil health and farmers’
wealth adversely.
41
CHAPTER-III
Survey Analysis and Design
3.1 IntroductionIn the previous chapter, A review of literature on agriculture crop yield
prediction based yield variables like a soil, weather, plant analysis and land
management etc was discussed. This chapter exhibits yield prediction of banana
based on different type’s of soil in India with a difference in the genetic and
environment factors. Plant analysis is the quantitative analysis of the total nutrient
content in a plant tissue. It is based on the principle that the amount of a nutrient in
diagnostic plant parts indicates the soil’s ability to supply that nutrient and is directly
related to the available nutrient status in the soil. Artificial neural network training is
used to get yield variables to get optimum result. Hence, it seems quite convenient
and appealing for bananas also.
3.2 Soils
Soil may be defined as a thin layer of earth’s crust which serves as a natural
medium for the growth of plants. It is the unconsolidated mineral matter that has been
subjected to, and influenced by genetic and environmental factors parent material,
climate, organisms and topography all acting over a period of time. Soil differs from
the parent material in the morphological, physical, chemical and biological properties.
Also, soils differ among themselves in some or all the properties, depending on the
differences in the genetic and environmental factors. Thus some soils are red, some
are black; some are deep and some are shallow; some are coarse-textured and some
are fine-textured. They serve in varying degree as a reservoir of nutrients and water
for crops, provide mechanical anchorage and favorable tilth. The components of soils
are mineral material, organic matter, water and air, the proportions of which vary and
which together form a system for plant growth hence the need to study the soils in
perspective8.
8 A study of soil profile supplemented by physical, chemical and biological properties of the soilwill give full picture of soil fertility and productivity. Department of Agriculture & CooperationMinistry of Agriculture Government of India New Delhi January, 2011
42
Rocks are the chief sources for the parent materials over which soils are
developed. There are three main kinds of rocks: (i) igneous rocks, (ii) sedimentary
rocks and (iii) metamorphic rocks.
The rocks vary greatly in chemical composition and accordingly the soil
differs in their properties because they are formed from the weathering of rocks.
Weathering can be physical or chemical in nature. The agents of physical weathering
are temperature, water, wind, plant and animals while chemical processes of
weathering are hydration, hydrolysis, carbonation, oxidation and reduction.
A developed soil will have a well defined profile which is a vertical section of
the soil through all its horizons and it extends up to the parent materials. The horizons
(layers) in the soil profile which may vary in thickness may be distinguished from
morphological characteristics which include colour, texture, structure etc. Generally,
the profile consists of three mineral horizons – A, B and C.
The A horizon may consist of sub-horizons richer in organic matter intricately
mixed with mineral matter. Horizon B is below A and shows dominance of clay, iron,
Aluminum and humus alone or in combination. The C horizon excludes the bedrock
from which A and B horizon are presumed to have been formed.
Physical properties of the soil include water holding capacity, aeration,
plasticity, texture, structure, density and colour etc. Chemical properties refer to the
mineralogical composition and the content of the type of mineral such as Kaolinite,
illite and montmorillonite, base saturation, humus and organic matter content. The
biological property refers to a content of extent and types of microbes in the soil
which include bacteria, fungi, worms and insects.
43
3.3 Major soil types of India
Some dominant groups of Indian soil, classified according to soil taxonomy and
chemical property are mentioned below:
3.3.1) Red soil
They are quite wide in their spread. The red colour is due to diffusion of iron
in the profile.
3.3.2) Lateritic soil
Lateritic soil is composed of a mixture of hydrated oxides of aluminum and
iron with small amounts of manganese oxide.
3.3.3) Black soil
Black soil contains a high proportion of Calcium and Magnesium Carbonates
and has a high degree of fertility.
3.3.4) Alluvial soils
This is the largest and agriculturally most important group of soils.
3.3.5) Desert soils
Desert soils occur mostly in dry areas and its important content is quartz.
3.3.6) Forest and Hill Soils high in organic matter
The soils are studied and classified according to their use which is termed
as land capability classification. In this classification, inherent soil characteristic,
external land features and environmental factors are given prominence. For this
purpose, soil survey is carried out to record the crop limiting factors such as soil
depth, topography, texture-structure, and water holding capacity, drainage features,
followed by evaluation of soil fertility status, based on soil testing / analysis.
3.4 Soil groups
The above soil groups which have been extensively studied because of their
extent and agricultural importance are described below:
44
3.4.1 Red Soils
The red soils of India, including red loams and yellow earths, occupy about
200,000 sq.miles and extend over a large part of Tamil Nadu, Mysore, south-east
Maharashtra and a tract along the eastern part of Madhya Pradesh to Chota Nagpur
and Orissa. In the north and north-east these extend into and include great part of the
Santhal Parganas of Bihar Birbhum, Bankura and Midnapur districts of West Bengal
Khasi, Jaintia, Garo and Naga Hills areas of Assam Mirzapur, Jhansi, Banda and
Hamirpur districts of Uttar Pradesh Baghelkhand division of Madhya Pradesh and
Aravallis and the eastern half of Rajasthan.
The main features of these soils, besides their lighter texture and porous and
friable nature, are: (a) the absence of lime (kankar) and free carbonates, and (b) the
usual presence of soluble salts in a small quantity, not exceeding 0.05 percent. These
soils are generally neutral to acid in reaction and deficient in nitrogen, phosphoric
acid, humus and perhaps lime. They differ greatly in depth and fertility, and produce
a large variety of crops under rain fed or irrigated conditions. They are divided into
two broad classes: (1) the red loams, characterized by a cloddy structure and the
presence of only a few concretionary materials; and (2) the red earth with loose top-
soil, friable but rich secondary concretions of a sesquioxidic clayey character.
The soil contains a high percentage of decomposable hornblende, suggesting a
comparatively immature nature. The silica-alumina ratio of the clay fractions is 2.7-
2.46 and their base exchange capacities are below 20 m.e. per 100 gm., suggesting
their predominantly kaolintic nature. In the typical red earth the silica-alumina ratio
of the clay fractions is higher than 2 and they are fairly rich in iron oxide.
The soils have undergone excessive weathering and very low amount of
decomposable mineral hornblende. In Tamil Nadu the red soils occupy a large part of
cultivated area. They are rather shallow, open in texture with the pH ranging between
6.6 and 8.0. They have a low base status and low exchange capacity, and are deficient
in organic matter, poor in plant nutrients, and with the clay fraction ratio of 2.5 – 3.0.
45
The predominant soil in the eastern tract of Mysore is the red soil, overlying
the granite from which it is derived. The loamy red soils are predominant in the
plantation districts of Shimoga, Hassan and Kadur. They are rich in total and
available K2O, and contain fair amounts of total P2O5 (0.5 – 0.3 percent); the lime
content is 0.1 – 0.8 per cent, nitrogen below 0.1 per cent, and iron and alumina 30 –
40 per cent. A broad strip of area lying between the eastern and western parts of
Coorg is red loam, easily drained and with a fairly dense growth of trees.
The acid soils in the south of Bihar (Ranchi, Hazaribagh, Santhal Parganas,
Manbhum and Singhbhum) are red soils. Their pH is 5.0 – 6.8 and they have high
percentage of acid-soluble Fe2O3 as compared with Al2O3 ; sufficient available
potash but P2O5 is low. The soils from Manbhum, Palamau and Singhbhum are
preponderant in zircon, hornblende and rutile respectively; those of Ranchi contain a
mixture of epidote and hornblende, neither of which is preponderating.
In West Bengal the red soils, sometimes misrepresented as laterites, are the
transported soils from the hills of the Chhota Nagpur Plateau. The existing tracts of
soils in north-west Orissa are quite heterogeneous. There seems to be a prominent
influence of the rolling and undulating topography on soil characteristics. The soils
are slightly acidic to neutral in reaction and the total soluble salts are fairly low.
Ferruginous concretions are invariably met with, whereas calcareous
concretions are present only in a few cases at lower depths of the profiles. In a typical
red soil profile the total exchangeable bases is about 20 m.e., the SiO2- R2O3 ratio of
the clay fractions varies between 2 and 3, and the C – N ratio is near about 10.
The soils of Raipur district (Chattisgarh area) are grouped into the following classes:
Dorsa
These are Soils along the slopes, somewhat darker with same texture as above and
good paddy lands.
Kanhar
Kanhar is Lowland soil, dark, slightly heavier than Dorsa paddy is the main
crop and wheat is also grown in these lands.
46
Bhate
These are barren waste lands with gravel and sandy reddish-yellow and
usually in uplands. A part of Jhansi district (Uttar Pradesh) comprises red soils. These
are of two types : Parwa, a brownish-grey soil, varying from good loam and sand or
clay loam, and rakar, the true red soil is generally not useful for cultivation.
In the Telungana division of Andhra Pradesh both red and black soils
predominate. The red soils or chalkas are sandy loam located at higher levels and are
utilized for cultivation of kharif crops.
Another type of soil occurring in Andhra Pradesh is locally known as dubba.
It is loamy sand or very coarse sandy loam, and mostly pale-brown to brown with
reddish-brown patches here and there ; clay content is quite low (less than 10 percent)
and it has very low fertility ; invariably neutral in reaction and low in soluble salt
content. The content of organic matter is little to negligible. The soils are severely
eroded with surface soil depth below five inches and very often covered with multi-
sized gravels and cobbles. Being sub-marginal lands they are well suited for pasture
and forage crops rather than for rice growing.
3.4.2 Laterite and lateritic soils:
These soils occupy an area of about 49,000 sq.miles in India. The laterite
is specially well-developed on the summits of the Deccan Hills, Central India,
Madhya Pradesh, the Rajmahal Hills, the Eastern Ghats, in certain plains of Orissa,
Maharashtra, Malabar and Assam. These are found to develop under fair amount of
rainfall, and alternating wet and dry periods. The laterite and lateritic soils are
characterized by a compact to vesicular mass in the subsoils horizons composed
essentially of mixture of the hydrated oxides of aluminium and iron. These soils are
deficient in potash, phosphoric acid and lime. On higher levels these soils are
exceedingly thin and gravelly, but on lower levels and in the valleys they range from
heavy loam to clays and produce good crops, particularly rice.
47
Both the high-level and low-level laterites occur in Tamil Nadu. They are both
in situ and sedimentary formations, and are found all along the West Coast and also in
some parts of the East Coast,where the rainfall is heavy and humid climate prevails.
In the laterites on lower elevations paddy is grown, while tea, cinchona, rubber and
coffee are grown on those situated on high elevations. The soils are rich in nutrients
including organic matter. The pH is generally low, particularly of the soils under tea
(pH 3.5 – 4.0) and at higher elevation.
In Maharashtra laterites are found only in Ratnagiri and Kanara; those in
the latter are coarse, poor in lime and P2O5, but fairly good in nitrogen and potash. In
the former, coarse material abounds in large quantities. These are rich in plant food
constituents, except lime.
In Kerala, between the broad sea belt consisting of sandy soil and sandy loams
and the eastern regions comprising the forest and plantation soils, the mainland
contains residual laterite. These are poor in total and available P2O5, K2O and CaO.
Laterite rock in Cochin is found to the east of the alluvial areas in Trichur, Talapalli
and Mukundapuram taluks. Soil is mostly laterite in Trichur taluk.
The nitrogen content varies from 0.03 – 0.33 per cent; the lime is very poor
and the magnesium is 0.11 – 0.45 per cent. The laterite soils in Mysore occur in the
western parts of Shimoga, Hassan, Kadur and Mysore districts. All the soils are
comparable to the laterites and to the similar formations found in Malabar, Nilgiris,
etc. These soils are very low in bases, like lime, due to severe leaching and erosion.
These are poor in P2O5. The pH is not as low as that in the plantation soils.
In West Bengal, the area between Damodar and Bhagirathi is interspersed
with some basaltic and granitic hills with laterite capping. Bankura district is known
to be located in the laterite soils zone. The SiO2 – Al2O3 ratio of the clay fraction is
quite high. The percentage of K2O, P2O5 and N are also low, showing considerable
leaching and washing out of these substances due to chemical weathering. The soils
of Burdwan are in all respects similar to the Birbhum and Bankura soils with one or
two exceptions. The high value of the SiO2 – Al2O3 ratio is significant.
48
In Bihar laterite occurs principally as a cap on the higher plateau but is also
found in some valleys in fair thickness. The laterites of Orissa are found largely
capping the hills and plateau occasionally in considerable thickness. Large areas in
Khurda are occupied by laterites. At Balasore, it is gravely. Two types of laterites are
found in Orissa, the laterite murrum and the laterite rock. They may occur together.
3.4.3 Black soils
These soils cover a large area throughout the southern half of the peninsula,
the Deccan Plateau, greater part of Maharashtra State, western parts of Madhya
Pradesh and Andhra Pradesh, and some parts of Tamil Nadu State, including the
districts of Ramnad and Tinnevelly. The black soils or regur include a large number
of physiographic regions, each within a zone having its own combination of soils.
These soils may be divided into three groups : (1) deep and heavy; (2) medium and
light; and (3) those in the valleys of rivers flowing through regur area.
The main features of the black soils are: (1) depth one to two or several feet
deep; (2) loamy to clayey in texture; (3) cracking heavily in summer, the cracks
reaching up to more than three or four feet in depth, especially in the case of heavy
clays; and (4) containing lime kankar and free carbonates (mostly CaCO3) mixed with
the soil at some depths. These soils are often rich in montmorillonitic and beidlite
group of minerals, and are usually suitable for the cotton cultivation. They are
generally deficient in nitrogen, phosphoric acid and organic matter; potash and lime
are usually sufficient. The content of water-soluble salts is high, but the investigations
carried out in connection with Tungabhadra and Nizamsagar projects have shown that
these soils may be irrigated without any danger, if irrigation is carried out on sound
lines.
Though the black soils do not have distinct demarcation of horizons between
the un weathered parent material and the weathered soil, the soil profile may be said
to possess approximately three principal horizons A, B and C, the alluvial or A
horizon being predominant and of two types, namely, darker with high organic matter
49
content and lighter. The zone of accumulation of carbonates (CaCO3) and sulphates
(chiefly CaSO4) may be taken as the B or illuvial horizon. In regions of fairly high
and evenly distributed rainfall the zone of carbonate accumulation is found deeper in
the profile and sometimes incorporated with horizon C.
The occurrence of black and red soils in close proximity is quite common in
India. In Maharashtra soils derived from the Deccan trap occupy quite a large area.
On the uplands and on the slopes, the black soils are light coloured, thin and poor;
and on the lowlands and the valleys they are deep and relatively clayey. Along the
Ghats the soils are very coarse and gravelly. In the valleys of the Tapti, the Narmada,
the Godavari and the Krishna heavy black soil is often 20 feet deep. The subsoil
contains good deal of lime. Outside the Deccan trap the black cotton soils
predominate in Surat and Broach districts. Degraded solonized black soils, locally
known as chopan, occur along the canal zones in the Bombay Deccan. A large
number of typical black soil profiles have been examined in Tamil Nadu. They are
either deep or shallow and may or may not contain gypsum in their profile, and
accordingly four types of profiles are distinguished: (1) shallow with gypsum, (2)
shallow without gypsum, (3) deep with gypsum, and (4) deep without gypsum. The
shallow profiles are three to four feet deep, often with partially weathered rock
material even at a depth of 1.5 – 2.0 feet; the deep ones extend even up to nine feet or
more.
The black soils are very heavy, contain 65-80 per cent of finer fractions, have
high pH (8.5 – 9.0) and are rich in lime (5-7 per cent); they have low permeability,
high values of hygroscopic coefficient, pore-space, maximum water-holding capacity
and true specific gravity. They are low in nitrogen but contain sufficient potash and
P2O5. They have generally a high base status and high base exchange capacity (4 60
meg.) ; about 10-13 per cent iron content, and the CaO and MgO contents are formed
from a variety of rocks, including traps, granites and gneisses.
In Madhya Pradesh the black soils are either deep and heavy (covering the
Narmada Valley) or shallow (in the districts Nimar, Wardha, west of Nagpur, Saugor
and Jabalpur). The cotton-growing areas are mainly covered by the deep heavy black
soils which, however, gradually change in colour from deep-black to light. The
50
CaCO3 content increases with the depth. Clay content is 35-50 per cent, the organic
matter is low and SiO2-R2O3 ratio is 3 – 3.5.
The black soils of Mysore are fairly heavy with high salt concentration, and
rich in lime and magnesia. The SiO2-R2O3 ratio of clay fraction is 3.6.
3.4.4 Alluvial soils:
The so-called alluvial soils of India form an ill-defined group. Various types
of alluvium are classed as alluvial, e.g., calcareous soils, saline and alkali soils, and
coastal soils. The alluvial soils occur mainly in the southern, north-western and north-
eastern parts of India: the Punjab, Uttar Pradesh, Bihar, West Bengal, parts of Assam,
Orissa, and coastal regions of southern India including the local deltaic alluvia. These
soils are the most fertile amongst the Indian soils. The whole of the Indo-Gangetic
plain is, in this alluvial area, of 300,000 square miles. These soils are very deep,
deeper than 300 ft. at some places, and deficient in nitrogen and humus, occasionally
in phosphoric acid but not generally in potash and lime. They support a variety of
crops, including rice, wheat and sugarcane. They may be sub-divided into two broad
groups, the old and the new; both are geological groupings. The former, locally called
bangar, represents reddish-brown, sandy loams with increasing content of clay in the
lower horizons ; the latter, known as khaddar, represents the fairly coarse sand on the
chars and banks of the river to the soils of very fine texture in the low-lying marshy
tracts. The old alluvium reddish in colour, is deficient in nitrogen and humus, and
occasionally in phosphoric acid.
The large expanse of these soils is yellowish to brownish and their common
feature is the presence of kankar or lime nodules intermixed with soil at varying
depths. They vary from sandy loam to clayey loam. The subsoil occasionally has
calcareous reaction. There is no marked differentiation into the various horizons, and
the profile is often characterized by the absence of stratification. The surface soil is
generally grey, varying from yellow to light brown, the intensity of colour increasing
with the depth. The immature soil near the rivers is calcareous and light brown in
colour with salt impregnation. On higher situations it becomes brown to deep brown
in colour and is non-calcareous. Kankar beds are found in the soil. Most of the
alluvial soils in Uttar Pradesh and Bihar are of the above pattern.
51
3.4.5 Desert Soil:
A large part of the arid region in Rajasthan and part of Haryana, lying
between the Indus and the Aravallis, is affected by desert conditions of recent
geological origin. This part is covered under a mantle of blown sand which inhibits
the growth of soils. The Rajasthan desert proper (area about 40,000 sq. miles), owing
to its physiographic conditions receive no rain though lying in the tract of the south-
west monsoon. Some of the desert soils contain high percentage of soluble salts, high
pH, varying percentage of calcium carbonate and poor organic matter, the limiting
factor being mainly water. The soils could be reclaimed if proper facilities for
irrigation are available.
3.4.6 Forest and hill soils:
Nearly 22-23 per cent of the total area of India is under forests. The formation
of forest soils is mainly governed by the characteristic deposition of organic matter
derived from the forest growth. Broadly two types of soil-formation may be
recognized (1) soils formed under acid conditions with presence of acid humus and
low base status; and (2) soils formed under slightly acid or neutral condition with
high base status which is favourable for the formation of brown earths. The soils of
the hilly districts of Assam are of fine texture and reveal high content of organic
matter and nitrogen, perhaps due to the virgin nature. Their chemical and mechanical
composition show great variations.
52
Soil testing refers to the chemical analysis of soils and is well recognized as a
scientific means for quick characterization of the fertility status of soils and predicting
the nutrient requirement of crops. It also includes testing of soils for other properties
like texture, structure, pH, Cation Exchange Capacity, water holding capacity,
electrical conductivity and parameters for amelioration of chemically deteriorated
soils for recommending soil amendments, such as, gypsum for alkali soils and lime
for acid soils. One of the objectives of soil tests is to sort out the nutrient deficient
areas from non-deficient ones. This information is important for determining whether
the soils could supply adequate nutrients for optimum crop production or not.
3.5 Plant Analysis as Nutritional Requirements of Bananas
Plant analysis has been considered as a very practical approach for
diagnosing nutritional disorders and formulating fertilizer recommendations (Kelling
et al., 2000[49]; Self, 2005). Plant analysis, in conjunction with soil testing, becomes
a highly useful tool not only in diagnosing the nutritional status but also an aid in
management decisions for improving the crop nutrition (Rashid, 2005)[80]. Plant
analysis is the quantitative analysis of the total nutrient content in a plant tissue, based
on the principle that the amount of a nutrient in diagnostic plant parts indicates the
soil’s ability to supply that nutrient and is directly related to the available nutrient
status in the soil (Malavolta, 1994[62]; Kelling et al., 2000[49]; Havlin et al.,
2004[37]; Rashid, 2005)[80]. It is a very practical and useful technique for fruit trees
and long duration crops (Rashid, 2005). Hence, it seems quite convenient and
appealing for bananas also.
Bananas are heavy feeder of nutrients (Jones, 1998)[47] and thus need
balanced nutrition for optimum growth and fruit production, and in turn potential
yields. A deficiency or excess of nutrients can cause substantial damage to the plant (
Memon et al., 2001)[68].
53
The early (until the mid-1960s) researches on banana nutrition had
concentrated on the description of symptoms of nutrient imbalance and the conduct of
field experiments comparing response to rates of applied fertilizer on a range of soil
types. To aid in determining the nutrient supplying power of the soil, aid in
determining the effect of treatment on the nutrient supply in the plant, study
relationship between the nutrient status of the plant and crop performance as an aid in
predicting fertilizer requirements, help lay the foundation for approaching new
problems or for surveying unknown regions to determine where critical plant
nutritional experimentation should be conducted. The succeeding research workers
opined almost similarly about the uses of plant analysis (Smith, 1986, Jones, et al.,
1991[44], Kelling et al., 2000[49]; Havlin et al., 2004[37]; Rashid, 2005; Self, 2005).
For plant analysis, a specific plant part at a particular growth stage should be
sampled because comparison of an assay result with established critical or standard
values or sufficiency ranges is used to interpret analytical results (Rashid, 2005). It is
important to follow the recommended sampling technique carefully, since criteria for
elemental analysis interpretation have been established for specific plant sampling
procedures. Therefore, for meaningful determinations of the elemental concentration,
it is essential to adhere to the given sampling procedure designed for that plant
species and the element(s) to be assayed (Jones, 1997)[46].
Sampling procedures have been investigated by many researchers (Dumas,
1959[25]; Twyford & Coulter, 1964;Martin-Prevel et al., 1969; Lahav, 1970[53];
Turner & Barkus, 1977). Earlier, researchers at the Jamaica Banana Board (Hewitt,
1953[39]; Hewitt & Osborne, 1962[40]) and IRFA, Guinea (Dumas & Martin-Prevel,
1958[24]; Dumas, 1960a[26]), used different approaches and defined some of the
problems associated with sampling in banana. It was thus difficult to perceive
indisputable overall advantage in either one method or the other and hence many
workers preferred to establish a procedure well suited to their own special
circumstances. In two decades, a variety of procedures were used. Later on, Martin-
Prevel (1977)[65] came up with a measure of uniformity to sampling methods by
surveying the methods used in different countries.
54
Because of the internal variation in nutrient composition of banana, the results
from these different techniques were, almost without exception, not strictly
comparable. Lahav and Turner (1983)[57] attributed the slow progress towards
international standardization of sampling techniques "partly to the nature of the
banana plant and partly to the absence of unifying concepts concerning its nutrition".
The interplay of growth and nutrition is more complex in the banana than
most crops and best understood from detailed data on the nutrient flux in the plant as
a whole. Realizing the need for uniformity of sampling method and to provide for
comparison of results between experiments conducted in different countries, the
International Working Group on Foliar Analysis in the Banana was established. The
Working Group met for the first time in 1975 in the Canary Islands. There was a
general realization of the advantages of standardization of sampling methodology.
Figure-1 shows the Sampling procedures for banana leaves (Martin-Prevel,1977)
The first outcome was that each organization agreed to standardize procedures
wherever this could be done without difficulty and to move towards an international
55
reference sampling method (Method d'Echantillonnage Internationale de Reference -
MEIR) (Martin-Prevel, 1974, 1976, 1977).Area of sampling. According to MEIR
samples are taken from three leaf parts at different positions on the plant (figure- 1).
The samples should normally be taken either just before or following floral
emergence and when all female hands are visible (Martin-Prevel, 1974; 1976; 1977;
Lopez & Espinosa, 2000). However, the age of the tissue to be sampled depends on
the nutrient being diagnosed (Lopez & Espinosa, 2000). For instance, sulphur is
better diagnosed if younger leaves are sampled before floral initiation (Fox et al.,
1979).
In most banana producing countries, the laminar structure of third leaf is sampled
for tissue analysis. However, samples of the central vein of third leaf and the petiole
of seventh leaf are also used. The laminar structure of third leaf is sampled by
removing a strip of tissue 10 cm wide, on both sides of the central vein, and
discarding everything but the tissue that extends from the central vein to the center of
the lamina (Lopez & Espinosa, 2000). The MEIR method allows for comparison of
results between experiments, but whether it is the best method for a diagnostic service
still remains to be established (Memon et al., 2001)[68].
Further developments in sampling methods and some of the unresolved issues
were reviewed in detail by Martin- Prevel (1980). He considered that the
development of a uniform method of sampling was slow, especially when the benefits
were considerable. Since the establishment of International Working Group and their
first meeting in 1975, there have been two enlarged meetings on the "Nutrition of
Banana Crop” in Australia in 1978 and on the “Agro-physiology of Bananas” in
South Africa in 1982.
Although considerable progress has been made in standardization, there is still
much to be done to achieve complete uniformity. Almost all the information on
assessment of nutrient status in the banana plant relates to leaf sampling – blade,
midrib or petiole. There have been a number of investigations on other organs to
quantify nutrient uptake or removal, only the leaf blade was considered in the first
wave of investigations. In view of its size, it was not practicable to take the whole leaf
56
as a sample. For that, Dumas (1960b)[27] mapped the spatial variability in the
mineral content of banana leaf blade, in an attempt to find areas of constant
composition and reasonable size. The variations within each half of the blade were
considerable, both transversely and longitudinally (Fig. 2).
As a result, whatever part of the blade was chosen it must be precisely
defined, and the analyses interpreted only by reference to norms based on data for
that part of the leaf. Lahav (1972a) [54]pointed out that a 5 cm longitudinal
displacement of the area sampled could give a difference in K content equivalent to
that from an application of K fertilizer. Specifications such as "in the middle of the
leaf" or about the first third of the leaf were inadequate. Variability between leaves is
somewhat less in the central part of the leaf than it is in the basal and distal areas (Fig.
2). This is one reason why most authorities have chosen to sample parts of the central
area rather than the extremities. Further work of Lahav (1972b[55], 1977[56])
revealed that petiole analysis provided more information than the blade, at least for
cations and phosphorus (P). Martin-Prevel et al. (1968) and Martin-Prevel (1970) also
showed that the conductive tissues were useful indicators for cations. They found it
best, however, to take the section of the midrib adjacent to the area of blade that was
already being sampled (Martin-Prevel et al., 1969). Langenegger and Du Plessis
(1977) reached a similar conclusion and have since re-emphasized their preference
for the midrib including its use to indicate plant nitrogen (N) status. Hewitt (1953)
[39]analyzed all odd numbered leaves and found that N content was highest at about
position III. He, therefore, chose this as a standard and was followed in doing so by
research groups in most countries. Position III has accordingly been adopted as the
international standard.
For a diagnostic service, the appropriate sampling method is one that allows an
empirical relation between the concentration of the nutrient and response to the
application of that nutrient to be established. It may be that a single sampling method
will not cater for all nutrients under all climatic and soil conditions (Lahav, 1972b;
1977)[55]. A full evaluation of the recommended sampling methods has yet to be
57
completed but indications are that the petiole or midrib may be better than lamina for
assessing P status.
The figure-2 shows the spatial variability in the mineral content of leaf blade of
banana cultivar Dwarf Cavendish.
Figures in upper part of leaf are mean nutrient content of n leaves as % of DM
and figures in the lower half are coefficient of variation of those means (Dumas,
1960b)[27].
58
3.5.1 Stage of sampling.
A further requirement for a sampling method is that the variation from plant to
plant within a tissue is as low as possible. Twyford and Walmsley (1974), who
sampled 10 plants, found that the usual diagnostic tissue used in the West Indies (the
fourth leaf lamina) was the least variable for all elements and all other plant parts,
especially at the "large" stage of plant growth. It is also important that the diagnostic
tissue, besides reflecting low plant-to-plant variability should indicate the nutrient
status of the whole plant. For example, Twyford and Walmsley (1974) found that the
concentration of potassium (K) in the leaves (3%) or petioles (3.2%) at the "large"
stage was the same for two sites in Windward Islands but at one site the plant
contained 210 g K and the other only 108 g K. Therefore, a quantitative estimate of
plant height, if used in conjunction with the concentration data, may give an estimate
of whole plant nutrient content.
According the international standard, (Martin-Prevel, 1980) sampling stage in
short banana plants is when all female hands are visible and up to 3 male or mixed
hands have formed. The appearance of three of the latter takes about a week, so that
the sampling period is a week long. The main advantage of this sampling stage is that
most of the current growth cycle is over, so that its effects are reflected in the sample
taken, yet there is opportunity to estimate yield and adequate time for interpretation
before the next cycle begins. The sampler can obtain a yield estimate by counting the
number of hands and of fingers per hand and also assess growth by measuring the
circumference of the pseudo stem at a standard height. Its disadvantage is a less
information on a standard nutrient contents and repeatability of the results at this
growth stage, which was little used before its adoption as an international standard
(Martin-Prevel, 1980; Lahav & Turner, 1983)[57].
When information is needed on banana plants before inflorescence emergence,
the proposed standard is "at about inflorescence initiation" in the expectation that a
better method of defining this stage will in due course become available. Lahav
59
(1972a) [54] studied the factors influencing the potassium content of the third leaf of
the banana sucker. He reported that the K content of the 3rd leaf varied considerably
along the length of the blade. Other factors that had a marked effect on the K content
were leaf orientation, time of day, shade, irrigation and plant age. In another study,
Lahav (1972b)[55] grew bananas in sand culture with 5 levels of K and analysed all
plant parts. The foliar sheaths, petiole and midrib were all good indicators of the K
status of the plant. He recommended the sampling of the petiole of the 7th leaf as it
also contained relatively high concentrations of Ca, Mg, Na and Cl. Langenegger and
Plessis (1977) attempted to determine the nutritional status of Dwarf cavendish
banana in South Africa. They analyzed various plant parts in fertilizer experiments
and surveys of commercial plantings. The two most promising tissues for foliar
analysis were a section of midrib (midrib 2/3) and also the corresponding lamina from
the leaf in position III sampled after flowering at a stage when two hermaphrodite
hands were visible. The midrib sample gave a rather better indication of N and K
status as affected by fertilizer.
3.5.2 Taking representative sample Besides the stage of sampling, it is important to obtain a sample that will
represent the plantation. In an average crop, a representative sample can usually be
obtained form 20 plants at a given stage of growth, though in some cases 10 are
enough. In case of field experiments, it is better to sample 10-20 suitable plants per
plot when the majority of the plants in the crop reach the defined growth stage. For
example, for a post flowering sample, ignore the first 30% of plants that flower,
sample the next 40% and ignore the final 30%.
3.5.3 Plant Analysis Interpretation Once plant samples have been analyzed for desired nutrients, the next
question is usually whether the values found are sufficient to prevent the plant
suffering from deficiency. For this purpose, it is necessary to interpret plant analysis
data. For the interpretation of plant analysis data, various systems have been proposed
and used as follows.
60
The critical level concept. For correct interpretation of tissue analysis, the
interpreter must be familiar with the relationship between dry matter accumulation
and nutrient concentration. The general relationship between nutrient concentration in
plant tissue and plant yield is shown in Figure. Yield is severely affected when a
nutrient is deficient, and when the nutrient deficiency is corrected, growth increases
more rapidly than nutrient concentration (Havlin, et al., 2004)[37].
The figure-3 shows Relationship between essential nutrient concentration and plant
growth or yield (Havlin et al., 2004)37].
Under severe deficiency, rapid increases in yield with added nutrient can
cause a small decrease in nutrient concentration. This is called Steenberg effect and
results from dilution of the nutrient in the plant by the rapid plant growth. When the
concentration reaches the critical range, plant yield is generally maximized. Nutrient
sufficiency occurs over a wide concentration range, wherein yield is unaffected.
Increases in nutrient concentration above the critical range indicate that plant is
absorbing nutrients above that needed for maximum yield. This Luxury consumption
is common in most plants. Elements absorbed in excessive quantities can reduce plant
yield directly through toxicity or indirectly by reducing concentrations of other
nutrients below their critical ranges (Brady & Weil, 2004[13], Havlin et al., 2004)
[37].
61
Plants that are severely deficient in an essential nutrient exhibit a visual
deficiency symptom show the figure. Plants that are moderately deficient exhibit no
visual symptoms, although yield potential is reduced. Added nutrients will maximize
yield potential and increase nutrient concentration in plant. The term luxury
consumption means that plants continue to absorb a nutrient in excess of that required
for optimum growth. This extra consumption results in an accumulation of the plant
nutrient without corresponding increase in growth. However, with higher crop yields,
a greater concentration of nutrients is required. When nutrient toxicity occurs plant
growth and yield potential decrease, increasing the nutrient concentration in the plant
(Havlin et al., 2004)[37].
The figure-4 shows the Relationship between nutrient concentration in plant
and crop yield. The critical nutrient range represents an economic loss in yield
without visual deficiency symptoms (Havlin et al., 2004)[37].
62
The Critical Nutrient Concentration (CNC) is commonly used in interpreting
plant analysis results and diagnosing nutritional problems (Fig. 3 and 4). The CNC is
located in that portion of the curve where the plant-nutrient concentration changes
from deficient to adequate; therefore, the CNC is the level of a nutrient below which
crop yield, quality, or performance is unsatisfactory. However, considerable variation
exists in the transition zone between deficient and adequate nutrient concentrations
which makes it difficult to determine an exact CNC. Consequently, it is more realistic
to use the Critical Nutrient Range (CNR), which is defined as that range of nutrient
concentration at a specified growth stage above which the crop is amply supplied and
below which the crop is deficient (Kelling et al., 2000[49]; Tisdale et al., 2002; Brady
& Weil, 2004[13]; Havlin et al., 2004[37]; Rashid, 2005)[80]. This concentration
range lies within the transition zone, a range in concentration in which a 0% to 10%
reduction in yield occurs, with 10% reduction in yield point specified as critical value
of the element (Havlin et al., 2004)[37]. In an interpretative concept developed by
Okhi (1987), the critical nutrient level is that nutrient concentration level at which a
10% reduction in yield occurs; this level is also defined as the Critical Deficient Level
(CDL). Similarly, the Critical Toxic Level (CTL) is the concentration level at which
toxicity occurs. Critical nutrient ranges have been developed for most of the essential
nutrients in many crops.
Leaf analysis values in banana have been traditionally interpreted using the
critical value approach, a diagnostic tool that considers each nutrient independently of
one another. Many experiments on banana have established critical levels for all
essential nutrients. These levels are quite consistent despite being generated in
different countries having a wide range of environmental conditions, and established
from experiments involving various cultural treatments and practices. This
information has helped determine the amount of fertilizer needed for correcting
specific problems. Ramaswamy and Muthukrishnana (1974) reported that a critical
level of 1.40% N was proved to be an optimal level in Robusta banana. Soil
application of 150 g/plant was fixed as critical level for maximising the yield. The
results obtained by Jambulingam et al. (1975)[42] suggested that leaf K should be
above 4.3% for optimum production. Later work by Arunachalam et al. (1976)[7]
63
showed that adequacy level of nutrients in banana leaf ranged from 3.18-3.43, 0.46-
0.54, 3.36-3.76, 2.3-2.4 and 0.25-0.28% for N, P, K, Ca and Mg, respectively.
Valsamma Mathew (1980) found that the nutrient status of third leaf at shooting
ranged from 1.33 to 2.08% for N, from 0.14 to 0.17% for P and from 2.05 to 2.76%
for K. In case of N, Kotur and Mustaffa (1984)[50] reported that a rate of 210 g
N/plant, corresponding to 3.51% leaf N, produced the highest yield of 44.8 t/ha.
Fernandez-Falcon and Fox (1985)[29] concluded that K level in the soil of
less than 2.26 meq/100 g, and in the leaf of less than 3.2%, reduced banana yields. A
nitrogen level in the leaf of less than 2.6% also limited yields. Adinarayana et al.
(1986)[1] observed that the mean potassium concentration (3.25%) in normal banana
leaves was much higher than that observed in potassium deficient leaves (1.25%).
According to Ray et al. (1988), a leaf content of 2.8% N, 0.52% P and 3.8% K at
shooting was a good indicator of satisfactory subsequent productivity of Robusta
banana.
Lahav & Turner (1992)[59] forwarded a summary of proposed critical levels
in different banana tissues (Table3). However, this concept has limitations. Stage of
growth greatly influences nutrient concentrations and unless the crop sample is taken
at proper time, the analytical results will be of little significance. Coupled with this,
considerable skill on the part of the analyst is needed to interpret the crop analysis
results in terms of the overall production conditions (Tisdale et al., 2002). Dumas and
Martin-Prevel (1958) pointed out that if nutrients are considered individually, values
equal to or higher than the critical level are not always associated with high yield or
values lower than the critical levels are not always related to low yield. In this case,
they proposed the use of ratio instead of concentrations as diagnostic norms.
64
The Table-1 given below suggested critical levels of nutrients in different
tissue of completely developed banana plants.
Nutrient Lamina(Leaf 3) Central Vein
(Leaf 3)
Petiole(Leaf 7)
N 2.6 0.65 0.4
P 0.2 0.08 0.07
K 3.0 3.0 2.1
Ca 0.5 0.5 0.5
Mg 0.3 0.3 0.3
Na 0.005 0.005 0.005
Cl 0.6 0.65 0.7
S 0.23 --- 0.35
Mn 25.0 80.0 70.0
Fe 80.0 50.0 30
Zn 18.0 12.0 08.0
B 11.0 10.0 08.0
Cu 9.0 7.0 05.0
Mo 1.5-3.2 - ---
65
3.6 Artificial Neural networks and its Applications
An Artificial Neural Network (ANN) is an information-processing paradigm
that is inspired by the way biological nervous systems, such as the brain, process
information. The key element of this paradigm is the novel structure of the
information processing system. It is composed of a large number of highly
interconnected processing elements (neurons) working to solve specific problems.
3.6.1 Use of neural networks
Either humans or other computer techniques can use neural networks, with
their remarkable ability to derive meaning from complicated or imprecise data, to
extract patterns and detect trends that are too complex to be noticed. A trained neural
network can be thought of as an "expert" in the category of information it has been
given to analyze.
3.6.2 Advantages
Adaptive learning: An ability to learn how to do tasks based on the data given
for training or initial experience.
Self-Organization: An ANN can create its own organization or representation
of the information it receives during learning time.
Real Time Operation: ANN computations may be carried out in parallel, and
special hardware devices are being designed and manufactured which take
advantage of this capability.
Fault Tolerance via Redundant Information Coding: Partial destruction of a
network leads to the corresponding degradation of performance. However,
some network capabilities may be retained even with major network damage.
66
3.6.3 A simple neuron
An artificial neuron (figure-5) is a device with many inputs and one output.
The neuron has two modes of operation, the training mode and the using mode. In the
training mode, the neuron can be trained to fire (or not), for particular input patterns.
In the using mode, when a taught input pattern is detected at the input, its associated
output becomes the current output. If the input pattern does not belong in the taught
list of input patterns, the firing rule is used to determine whether to fire or not.
A simple neuron
Figure -5 a simple neuron
3.6.4 Sophisticated Neuron
A more sophisticated neuron (figure-6) is the McCulloch and Pitts model
(MCP). The difference from the previous model is that the inputs are 'weighted', the
effect that each input has at decision-making is dependent on the weight of the
particular input. The weight of an input is a number which when multiplied with the
input gives the weighted input. These weighted inputs are then added together and if
they exceed a pre-set threshold value, the neuron fires. In any other case the neuron
does not fire.
67
Figure 6 shows an MCP neuron
In mathematical terms, the neuron fires if and only if;
X1W1 + X2W2 + X3W3 + ... > T
The addition of input weights and of the threshold makes this neuron a very flexible
and powerful one. The MCP neuron has the ability to adapt to a particular situation by
changing its weights and/or threshold. Various algorithms exist that cause the neuron
to 'adapt', the most used ones are the Delta rule and the back error propagation. The
former is used in feed-forward networks and the latter in feedback networks.
Figure -7 shows architecture of Artificial Neural Networks
68
This is diagram represents an architecture of Artificial Neural Networks. The
input layer consists of measured variables or inputs fed into input nodes. Various
weights are attached to the inputs that determine how the inputs interact, and the sum
of the inputs passes through a hidden layer where network perform problem specific
sub functions and reaches an output value. This output is compared to a known
outcome, and the process is repeated using new weights in an effort to get closer to
the outcome. The end result of a neural network is an accurate predictive model.
3.6.5 Applications
The utility of artificial neural network models lies in the fact that they can be
used to infer a function from observations. This is particularly useful in applications
where the complexity of the data or task makes the design of such a function by hand
impractical.
Real-life applications on Artificial Neural Networks
Function approximation, or regression analysis, including time series prediction,
fitness approximation and modeling.
Classification, including pattern and sequence recognition, novelty detection and
sequential decision making.
Data processing, including filtering, clustering, blind source separation and
compression.
Robotics, including directing manipulators, Computer numerical control.
69
Application areas include system identification and control (vehicle
control, process control), quantum chemistry, game-playing and decision making
(backgammon, chess, poker), pattern recognition (radar systems, face
identification, object recognition and more), sequence recognition (gesture,
speech, handwritten text recognition), medical diagnosis, financial applications
(automated trading systems), data mining (or knowledge discovery in databases,
"KDD"), visualization and e-mail spam filtering.
Banana production systems at the current level of yields are not found to be
sustainable, in the long run, as there is significant depletion of plant nutrients in
soil. Build up and maintenance of soil fertility and consequent provision of
balanced nutrition to banana crop is key to sustain long term banana productivity.
70
3.7 Summary
In this analysis elaborate information on major soil types of India is given
along with their composition, plant nutrient and their functions, typical deficiency
symptoms of nutrients in plants, apart from procedure of sample collection and
methods of analysis. Detailed information has been provided about the establishment
of soil testing laboratories, basic cares required in the laboratories, calibration
procedures for testing methods and the need and procedures for collaborating with
Soils Research Institutes in ICAR system and concerned State Agricultural
Universities. Information about the usefulness of soil testing kit and mobile soil
testing vans along with their limitations and usefulness has been provided. Since the
soil testing laboratories are invariably required to analyse irrigation water samples,
hence a chapter on irrigation water analysis has been provided.
Tissue testing is the determination of the amount of a plant nutrient in the sap
of the plant, a semi-quantitative measurement of the unassimilated, soluble content. A
large amount of an un assimilated nutrient in the plant sap indicates that the plant is
getting enough of the nutrient being tested for good growth. If the amount is low,
there is a good chance that the nutrient is either deficient in the soil or is not being
absorbed by the plant because of lack of soil moisture or some other factors.
So in this work a systematic approach has been developed to train Artificial
Neural Networks based banana yield prediction with new model. To achieving this
target the model has developed the ANN Absolute Update Technique.
71
CHAPTER-IV
METHODOLOGY
4.1 INTRODUCTION
The previous chapter dealt with soil levels in India and plant analysis at
various level. This chapter discusses about architecture of the Absolute Update
Technique, prototype model of this technique and its various performances based on
soil properties and plant analysis (leaf nutrients) which is prediction of banana.
Banana is an important fruit crop of tropical and sub-tropical regions of the
world. It requires high quantity of nutrients that must be supplied through fertilization
to obtain optimum yields. Mining of nutrients from soil is a major problem causing
soil degradations and threatening long-term food production in developing countries.
This research is taken up for carrying out nutrient resources, which includes the
calculation of nutrient balance at micro (field) and macro (farm) level and evaluation
of trends in nutrient mining.
In a densely populated country like India, agricultural research mainly focuses
on increasing the problem during the green revolution area. The overall performance
in food grain production is encouraged by green revolution. It propells India towards
self-sufficiency in food production. The following approach to reduce nutrient loss
and increase effective fertilizer usage.
72
4.2 Nutrient requirement of a banana crop
For high yield of quality fruit, bananas require relatively large amounts of
nutrients as they extract considerable quantities of nutrients from the soil. Sustainable
fertilizer practices aim to maintain soil fertility.
A prototype model for Yield prediction for banana
Figure-8 shows the prototype model of banana yield prediction
This prototype model shows the aim of this research work. This model has three
constrains soil nutrients, leaf nutrients and cost-effective.
INTEGRATED NUTRIENT TAILORING
Soil nutrients Leaf nutrients Cost efficitive
73
4.3 Absolute Update Technique for ANN Banana Yield Prediction
The ANN based Absolute Update Technique is a new challenging
environment method. The process is done in three layers. Primarily, basic input
values are taken into account and in the second layer Absolute Update Technique
recommends adjustments in values. These values are combined with the input values
to produce the accepted optimum result which is already set in the third layer as
output values. Thus Artificial Neural Network can be trained to find out accuracy in
crop yield prediction.
4.3.1 Structure of Absolute Update Technique
It is a user friendly Artificial Neural Network based model for monitoring
nutrient flows and stock especially in soils and leaf. Absolute Update Technique
enables the assessment of trends based on the local knowledge on soil fertility
management and the calculation of nutrient balances. Utilizing these results one can
easily identify the factors limiting crop production in the farm or region and propose
possible solutions for adoption and testing. Absolute Update Technique is a tool
encompassing a well structured, a database and two simple parameters soil based and
leaf. Finally, a user- interface facilitates data entry and extraction of data from the
database to produce input for both models. The tool calculates flows and balances of
the macronutrients and micronutrients N, P and K through independent assessment of
major inputs and outputs using the following equation.
74
Absolute update technique for Indicating the application of ANN for BananaYield Prediction
An implementation of Artificial Neural Networks based Absolute Update
Technique
Figure-9 shows Absolute Update Technique based ANN
Collection of Data
Training the network of line
Storing the final weights
Train/update thenetwork once again
Display the prediction value
Recommended data with stored weights
Input the soil content as variables
If the output inthe Specified Range
Stop
Yes
no
75
The following procedure are involved in Absolute Update Technique
Step 1. Randomly selected input values are applied to the network on Input layer
Step 2. Randomly initialize the weight value assigned to all neurons in hidden layers.
Step 3. To assign the recommended value taken from input layer through hidden layer
Step 4. The output layer is shows the best match it is chosen as optimum output.
Step 5. Updating the neighboring nodes to same process and getting the exact node to
produce the optimum result with iteratively
Step 5. Goto step 1.
Step 6. Steps 1 to 5 are repeated for all input nodes.
Setp 7. Stop
76
4.3.2 Architecture of Absolute Update Technique Design
Input layer Hidden Layer Output
Soil N
Soil P
Soil K
Soil Organic mater
Soil In Organic mater
Leaf N
Leaf P
Leaf K
Leaf Organic mater
Soil In Organic mater
.
.Fertilizer N .
Fertilizer P
Fertilizer K
Cost Effective
Figure-10 shows Absolute Update Technique Architecture
yield
AbsoluteUpdateTechnique
AbsoluteUpdateTechnique
AbsoluteUpdateTechnique
77
This Architecture design shows the Feed-Forward, Artificial Neural network
using Absolute Update Technique for calculating Banana yields with soil properties
and leaf nutrients. The network training use data from the Morrow Plots. The input
factors assumes to influence Banana yield and for which data are available from the
Morrow Plots included:
Soil: Nitrogen, Phosphorus, Potassium,Organic and In-Organic matter
Leaf: Nitrogen, Phosphorus, Potassium, Calcium, Magnesium, Iron, Zing, Organic
and In-Organic matter
Management: Soil Nutrients, Leaf Nutrients, Yield
The segmentation of nutrients during the growing period is based on the
planting date. Many soil nutrients and leaf nutrients are compared and the final model
is include with the elements in the input vector and the elements in the hidden layer
and one element (banana yield), in the output vector. The transfer function for each
neuron in the hidden layer and output layer was the sigmoid function.
The accuracy of the trained ANN is evaluated by calculating and individual
modeling error for each of examples reserved for testing. Individual errors were
calculated as
Yield prediction error = predicted yield –actual yield X 100 %
Actual yield
Using this equation positive errors indicate over-predictions, while negative
errors indicate under-predictions. To obtain an overall accuracy measure of the test
samples, the RMS error is calculated as :
N
RMS error = prediction error 2
N
78
4.4 Soil Analysis4.4.1 Data Source
The Morrow Plots data is used to build up the ANN model. The Morrow Plots
is located on the campus of the National Research Centre for Banana
(NRCB),Thayanur post, Thogamalai Road Tirurapalli-602102.
TECHNICAL PROGRAMME
Crop : Banana
Varieties : All varieties
Soil : Alluvial (Typic Ustopept)
Creation of fertility gradient in the soil of experimental field
Figure-11 shows morrow plots taken in the experiment field
79
Creation of fertility gradient in the soil of experimental field
Figure-12 shows morrow plots taken in the experiment field
A level field of about 1 hectare which has low to medium level of soil fertility
and representative of the experimental station or area is to be chosen. The field is
divided into four equal strips and each strip into four equal plots. Soil samples are
collected from each plot from 0-30cm and 30-60 cm depth and analyzed for available
N, P and K status.
11 22 33 44
80
4.4.2 Creation of fertility gradient in the soil of experimental field
Figure-13 shows morrow plots soil experimental field
1 NPK – 100:50:200 kg/hectare
11 22 33 44
00 NNPPKK ½½ NNPPKK 11 NNPPKK 22 NNPPKK
Banana
BananaBanana
Banana
Banana
Banana
Banana
Banana
Banana Banana
Banana
Banana Banana
Banana Banana
Banana
81
The first strip receives no fertilizer (NPK), the second, third and the fourth
receive half (NPK), one (NPK) and two times (NPK) a standard dose of N, P and K
respectively.The fertilizer treatment combinations from 4 X 3 X 5 levels of N, P and
K including absolute controls were randomly allotted in each of the four strips and the
suckers of test crop, banana (Nendran and Rasthali) were planted. By substituting the
required parameters in the above equation, the fertilizer doses are arrived at for
desired yield targets of crops for a range of soil test values.
The experiment which was conducted by the National Research Centre for
Banana (NRCB), Trichy with Alluvial soil type and Rasthali ,Nendran banana
varieties is taken for a comparative study. The soil samples are taken with the given
sample plots to identify the targeted banana yield ratio.
Example Varieties : Rasthali and Nendran
Soil : Alluvial (Typic Ustropept, mixed, hyperthermic)
Treatments : Factor 1 Factor 2 Factor 3
N0 – no N P0 – no P K0 – no K
N50 – 50% rec. N P50 – 50% rec. P K50 – 50% rec. K
N100 – 100% rec. N P100 – 100% rec. P K100 – 100% rec. K
N150 – 150% rec. N K150 – 150 % rec. K
K200 – 200% rec. K
Replication : Three
Number of plants : Eight
*(Rec. NPK – 200:50:400 g/plant)
82
4.4.3 The selected Treatment Plots in the Combination
Table -2 show The Selected 24 treatment combinations out of 60 (4x3x5) possible
combinations
N0P0K0 N100P100K100 N100P100K200
N150P100K200 N0P0K50N150P0K150
N50P100K50 N0P0K100 N50P50K50
N150P0K100 N50P50K100 N0P100K100
N50P0K100 N100P0K0 N150P0K0
N0P100K0 N0P0K200 N0P0K150
N50P100K200 N100P0K100 N50P50K200
N0P0K200 N50P50K0 N0P0K0
This treatment recommended for banana yield prediction Ratio is N PK-
200:50:400 gm/plant.
83
4.5 Leaf Analysis
Leaf analysis values in banana have been traditionally interpreted using the
critical value approach, a diagnostic tool that considers each nutrient independently of
one another. Many experiments on banana have established critical levels for all
essential nutrients.
These levels are quite consistent despite being generated in different
countries having a wide range of environmental conditions, and established from
experiments involving various cultural treatments and practices. This information has
helped determine the amount of fertilizer needed for correcting specific problems.
Ramaswamy and Muthukrishnana (1974) reported that a critical level of
1.40% N was proved to be an optimal level in Robusta banana. Soil application of
150 g/plant was fixed as critical level for maximising the yield.
The results obtained by Jambulingam et al. (1975) suggested that leaf K
should be above 4.3% for optimum production. Later work by Arunachalam et al.
(1976) showed that adequacy level of nutrients in banana leaf ranged from 3.18-3.43,
0.46-0.54,3.36-3.76, 2.3-2.4 and 0.25-0.28% for N, P, K, Ca and Mg,respectively.
Valsamma Mathew (1980) found that the nutrient status of third leaf at
shooting ranged from 1.33 to 2.08% for N, from 0.14 to 0.17% for P and from 2.05 to
2.76% for K. In case of N, Kotur and Mustaffa (1984) reported that a rate of 210 g
N/plant, corresponding to 3.51% leaf N, produced the highest yield of 44.8 t/ha.
Fernandez-Falcon and Fox (1985) concluded that K level in the soil of less than 2.26
meq/100 g, and in the leaf of less than 3.2%, reduced banana yields. A nitrogen level
in the leaf of less than 2.6% also had limited yields.
84
Adinarayana et al. (1986) observed that the mean potassium concentration
(3.25%) in normal banana leaves was much higher than that observed in potassium
deficient leaves (1.25%). According to Ray et al. (1988), a leaf content of 2.8% N,
0.52% P and 3.8% K at shooting was a good indicator of satisfactory subsequent
productivity of Robusta banana.
Lahav & Turner (1992) forwarded a summary of proposed critical levels in
different banana tissues. However, this concept has limitations. Stage of growth
greatly influences nutrient concentrations and unless the crop sample is taken at
proper time, the analytical results will be of little significance. Coupled with this,
considerable skill on the part of the analyst is needed to interpret the crop analysis
results in terms of the overall production conditions (Tisdale et al., 2002).
Dumas and Martin-Prevel (1958) pointed out that if nutrients are considered
individually, values equal to or higher than the critical level are not always associated
with high yield or values lower than the critical levels are not always related to low
yield. In this case, they proposed the use of ratio instead of concentrations as
diagnostic norms.
For this analysis, the following table suggested critical levels of nutrients in different
tissue of completely Developed banana plants.
85
Figure-14 shows the sampling procedure of banana leaf
86
4.6 Summary
As it is discussed earlier, Crop yield history suggests that crop
production systems are very complex. Presently in agriculture Fertility Gradient
approach is used which speak on the analysis of past data only. It does not have any
relevancy for future prediction. Old vegetative approach gives single iteration result
only. The process of Fertility Gradient approach seems to be a much longer and
complicated process giving insufficient details to farmers in terms of accuracy in
finding the prediction of yield.
When compared with Old Vegetative methods, the ANN based Absolute
Update Technique is a modern and challenging environment method. This method is
more beneficial to the farmers and agricultural scientists for it brings in better and
accurate results in crop yield.
The Old vegetative method obtains crop yields with a minimum of accuracy
only, whereas the ANN based Absolute Update Technique shows absolute of
accuracy.
87
CHAPTER -V
RESUST AND DISCUSSION
5. 1 INTRODUCTION
The previous chapter dealt with soil and leaf nutrients is performed at various
levels with Absolute Update Technique. This chapter discusses about Absolute
Update Technique and its various performances based on soil properties and plant
analysis. This chapter examines the sample values such as soil initial test values and
leaf nutrients values form National Research Centre Banana (NRCB) trichy and give
the best suggestion for prediction of banana.
Number of nodes in the hidden layer of the network which represent the
values. As the number of experimental values increases, the number of nodes in the
hidden layer also increases. Due to this, the network may some times report or may
not report besides increasing the computational effort. Having seen all old vegetative
methods, ANN technique can be considered to be mere beneficial to the farmers.
There by the chapter probes into now this new method functions and brings out better
profitable results
88
5.2 Absolute Update Technique using Soil Test
The ANN based Absolute Update Technique is set in such way there are
The configuration of the network has to be fixed. The number of nodes in the
hidden layer can be the same or different. The final weights which are obtained from
this network are taken down. This network is further experimented till the desired
prediction performance of the network is obtained. The sample data collected form
National Research Center for Banana (NRCB) is given below
Soil Test based Result
This Table-3 shows the initial soil test values available N P K gram per plant
N P K NitrogenGram per plant
(N)
PhosphorousGram per plant
(P)
PotassiumGram per plant
(K)
250 30 40 167.5 33.0166666666667470.4
200 20 30 134 28.0666666666667 376.933333333333
220 25 35 147.4 29.6333333333333 414.013333333333
210 30 25 81.1 14.85 281.933333333333
220 28 36127.533333333333 24.76
375.093333333333
200 30 35 163.829.3833333333333 433.32
250 28 33117.833333333333 24.76 376.0133333333
This table shows the sample values collected from NRCB with which the initial soil
test were done to view the nutrients (NPK) level in each plant in grams.
89
This Chart shows the initial soil test values available N P K gram perplant
This Chart shows the sample values collected from NRCB with which the
initial soil test were done to view the nutrients (NPK) level in each plant in grams.
90
Need for NPK Nutrients for kg per hectare
N P K N/kg/hec P/kg/hec k/kg/hec
250 30 40 502.5 99.05 1411.2
200 20 30 402 84.2 1130.8
220 25 35 442.2 88.9 1242.04
210 30 25 243.3 44.55 845.8
220 28 36 382.6 74.28 1125.28
200 30 35 491.4 88.15 1299.96
250 28 33 353.5 74.28 1128.04
This table-4 shows the sample values collected from NRCB to find out the nutrients
(NPK) level in Kilogram per hectares.
91
This Chart shows the sample values collected from NRCB to find
out the nutrients (NPK) level in Kilogram per hectares.
92
N P K Available in the Fertilizers
N P K N-Avil-FertP-Avil-
Fert K-Avil-Fert
250 30 40 1092.39130434783 619.0625 2352
200 20 30 873.913043478261 526.25 1884.66666666667
220 25 35 961.304347826087 555.625 2070.06666666667
210 30 25 528.913043478261 278.4375 1409.66666666667
220 28 36831.739130434783 464.25 1875.46666666667
200 30 35 1068.26086956522 550.9375 2166.6
250 28 33 768.478260869565 464.25 1880.06666666667
This table-5 above gives details about nutrients levels available in
the fertilizer range.
93
This Chart gives details about nutrients levels available in the fertilizer range.
94
Fertilizer ratio in plant
C-Urea
C-Sup-Phos
C-Mut-Phos
A-Ur-Hec A-Su-Pho-Hec
A-Mut-Phos-Hec
350 300 450
7646.73913043481 3714.375 21168
320 290 430
5593.04347826087 3052.25 16208.1333333334
340 285 435
6536.86956521739 3167.0625 18009.58
345 285 390
3649.5 1587.09375 10995.4
350 300 450
5822.17391304348 2785.5 16879.2
This table-6 shows the rate of fertilizer and the ratio of application
to per plant in hectare.
95
This Chart shows the rate of fertilizer and the ratio of application to per
plant in hectare.
96
Expenses and Target and Profit
Total ExpensesFertilizer
Banana-Kg-Rate
Target Gross-Profit Net-Profit
32529.1141304348 7 25 175000 142470.885869565
24853.4268115943 6 20 120000 95146.5731884057
27713.5120652174 6 22 132000 104286.487934783
16231.99375 7 15 105000 88768.00625
25486.8739130435 7 20 140000 114513.126086957
This table-7 shows the levels of expenditure, excepted target and the actual
gross profit and net profit achieved by the farmer.
97
This Chart shows the levels of expenditure, excepted target and the
actual gross profit and net profit achieved by the farmer.
98
5.3 ABSOLUTE UPDATE TECHNIQUE USING LEAF NUTRIENTS
Table 8 Leaf Nutrients based result
Combination -1
Leaf N (%)? 2.1 Leaf N/P 5 Leaf P/K 0.144828 Leaf K/Ca 3.866667 f(N/P) -14.0935 f(P/K) 13.48208 f(K/Ca) -14.7659
Leaf P (%)? 0.42 Leaf N/K 0.724138 Leaf P/Ca 0.56 Leaf K/Mg 11.6 f(N/K) -6.08421 f(P/Ca) -2.80E-14 f(K/Mg) -18.3115
Leaf K (%)? 2.9 Leaf N/Ca 2.8 Leaf P/Mg 1.68 Leaf K/S 14.5 f(N/Ca) -25.7767 f(P/Mg) -5.6079 f(K/S) 21.63
Leaf Ca (%)? 0.75 Leaf N/Mg 8.4 Leaf P/S 2.1 Leaf K/Fe 193.3333 f(N/Mg) -27.9964 f(P/S) 59.41771 f(K/Fe) 906.7661
Leaf Mg (%)? 0.25 Leaf N/S 10.5 Leaf P/Fe 28 Leaf K/Zn 1611.111 f(N/S) 12.44245 f(P/Fe) 255.3462 f(K/Zn) 9590.991
Leaf S (%) 0.2 Leaf N/Fe 140 Leaf P/Zn 233.3333 Leaf K/Mn 147.9592 f(N/Fe) 83.95193 f(P/Zn) 5369.587 f(K/Mn) 1157.64
Leaf Fe (ppm) 150 Leaf N/Zn 1166.667 Leaf P/Mn 21.42857 Leaf K/Cu 4833.333 f(N/Zn) 4259.563 f(P/Mn) -7.02083 f(K/Cu) 98491.74
Leaf Zn (ppm) 18 Leaf N/Mn 107.1429 Leaf P/Cu 700 f(N/Mn) 976.3724 f(P/Cu) 7082.123
Leaf Mn (ppm) 196 Leaf N/Cu 3500 f(N/Cu) 20054.09
Leaf Cu (ppm) 6
99
This chart shows leaf nutrients ratio using absolute update technique.
Combination 1
-20000
0
20000
40000
60000
80000
100000
120000
Leaf Nutrients Ratio
Valu
es
Leaf N (%) Leaf P (%) Leaf K (%) Leaf Ca (%) Leaf Mg (%)Leaf S (%) Leaf Fe (ppm) Leaf Zn (ppm) Leaf Mn (ppm) Leaf Cu (ppm)
100
Table 9 Shows the Leaf Nutrients ratio using absolute update technique
Combination 2
Leaf N (%) 1.9 Leaf N/P 4.75 Leaf P/K 0.16 Leaf K/Ca 3.33333 f(N/P) -18.21 f(P/K) 27.278 f(K/Ca) -37.28Leaf P (%) 0.4 Leaf N/K 0.76 Leaf P/Ca 0.53333 Leaf K/Mg 12.5 f(N/K) 2.2714 f(P/Ca) -6.361 f(K/Mg) -7.062Leaf K (%) 2.5 Leaf N/Ca 2.53333 Leaf P/Mg 2 Leaf K/S 12.5 f(N/Ca) -38.49 f(P/Mg) 19.227 f(K/S) 3.605
Leaf Ca (%) 0.75 Leaf N/Mg 9.5 Leaf P/S 2 Leaf K/Fe 178.571 f(N/Mg) -12.49 f(P/S) 48.994 f(K/Fe) 435.66Leaf Mg (%) 0.2 Leaf N/S 9.5 Leaf P/Fe 28.5714 Leaf K/Zn 1666.67 f(N/S) 0.5925 f(P/Fe) 278.62 f(K/Zn) 10611Leaf S (%) 0.2 Leaf N/Fe 135.714 Leaf P/Zn 266.667 Leaf K/Mn 131.579 f(N/Fe) 38.865 f(P/Zn) 7327.1 f(K/Mn) 694
Leaf Fe (ppm) 140 Leaf N/Zn 1266.67 Leaf P/Mn 21.0526 Leaf K/Cu 5000 f(N/Zn) 5696.1 f(P/Mn) -12.73 f(K/Cu) 105336Leaf Zn (ppm) 15 Leaf N/Mn 100 Leaf P/Cu 800 f(N/Mn) 681.4 f(P/Cu) 8934.2Leaf Mn (ppm) 190 Leaf N/Cu 3800 f(N/Cu) 23916
Leaf Cu (ppm) 5
101
This chart shows leaf nutrients ratio using absolute update technique.
Combination 2
-20000
0
20000
40000
60000
80000
100000
120000
1 2 3 4 5 6 7 8 9 10 11 12 13
Leaf Nutrients Ratio
Valu
es
Leaf N (%) Leaf P (%) Leaf K (%) Leaf Ca (%) Leaf Mg (%)Leaf S (%) Leaf Fe (ppm) Leaf Zn (ppm) Leaf Mn (ppm) Leaf Cu (ppm)
This chart shows leaf nutrients ratio using absolute update technique.
102
Combination 3
Leaf N (%) 1.8 Leaf N/P 3.6 Leaf P/K 0.227273 Leaf K/Ca 3.142857 f(N/P) -44.487 f(P/K) 88.4458 f(K/Ca) -47.172
Leaf P (%)? 0.5 Leaf N/K 0.818182 Leaf P/Ca 0.714286Leaf
K/Mg 11 f(N/K) 15.4871 f(P/Ca) 35.0522 f(K/Mg) -26.834Leaf K (%)? 2.2 Leaf N/Ca 2.571429 Leaf P/Mg 2.5 Leaf K/S 11 f(N/Ca) -36.509 f(P/Mg) 57.6812 f(K/S) -10.905
Leaf Ca (%)? 0.7Leaf
N/Mg 9 Leaf P/S 2.5 Leaf K/Fe 162.963 f(N/Mg) -19.068 f(P/S) 101.114 f(K/Fe) -63.206Leaf Mg (%)? 0.2 Leaf N/S 9 Leaf P/Fe 37.03704 Leaf K/Zn 1833.333 f(N/S) -5.5991 f(P/Fe) 623.38 f(K/Zn) 13672.5
Leaf S (%) 0.2 Leaf N/Fe 133.3333 Leaf P/Zn 416.6667Leaf
K/Mn 122.2222 f(N/Fe) 13.8167 f(P/Zn) 16136.2 f(K/Mn) 429.162
Leaf Fe (ppm) 135 Leaf N/Zn 1500LeafP/Mn 27.77778 Leaf K/Cu 5500 f(N/Zn) 9048.01 f(P/Mn) 83.6915 f(K/Cu) 125870
Leaf Zn (ppm) 12Leaf
N/Mn 100 Leaf P/Cu 1250 f(N/Mn) 681.396 f(P/Cu) 17268.5Leaf Mn
(ppm) 180 Leaf N/Cu 4500 f(N/Cu) 32926.7Leaf Cu (ppm) 4
103
This chart shows leaf nutrients ratio using absolute update technique.
Combination 3
-20000
0
20000
40000
60000
80000
100000
120000
140000
Leaf Nutrients Ratio
Valu
es
Leaf N (%) Leaf P (%)? Leaf K (%)? Leaf Ca (%)? Leaf Mg (%)?Leaf S (%) Leaf Fe (ppm) Leaf Zn (ppm) Leaf Mn (ppm) Leaf Cu (ppm)
104
Final Yield constrains on Leaf nutrients
Nitrogen(N) Phosphorous(p) Potassium(K)2812.497 1420.158 12236.483362.679 1848.286 13000.784731.081 3826.502 15524.39
Table-11 shows leaf nutrients ratio using Absolute Update Technique
This table accumulates the different nutrients levels seen in the leaf based on
the application of various combinations of nutrient components as listed out in the
Table- 8, Table-9 and Table-10.
0
2000
4000
6000
8000
10000
12000
14000
16000
18000
Nitrogen(N) Phosphorous(p) Potassium(K)
Nitrogen (N)Phosphorus (p)Potassium (K)
This shows accumulates the different nutrients levels seen in the leaf based
on the application of various combinations of nutrient components as listed out in
the Table- 8, Table-9 and Table-10.
105
5.4 Discussion
The experiment which was conducted by the National Research Centre for
Banana (NRCB), Trichy with Alluvial soil type and Rasthali, Nendran banana
varieties is taken up there for a comparative study. The soil samples are taken with
the given sample plots to identify the targeted banana yield ratio. In this present scope
the research has implemented Absolute Update Technique to achieve the optimum
banana yield based on soil properties and leaf nutrients. The result shows in the
following figure-15, figure-16, figure-17, figure-18 and figure-19.
106
Figure -15 shows the nutrients levels in the banana plant
107
The figure shows the effect of different levels of NPK on Nendran
Bunches. To achieve this target Absolute Update Technique involves
various levels, based on initial soil test values and plant analysis (leaf
nutrients).
Figure -16 shows the effect of nutrients levels of NPK on Nendran
banana plant
108
The figure shows the effect of different levels of NPK on Rasthali Bunches.
To achieve this target Absolute Update Technique involves various levels, based on
initial soil test values and plant analysis (leaf nutrients).
Figure -17 shows the effect of nutrients levels of NPK on Rasthali
banana plant
109
This figure shows the results of Absolute Update Technique. The results
exhibit the difference in the weight of the bunches depending on the combinations of
nutrients.
Figure -18 shows the effect of nutrients levels of NPK on
banana plant
110
The figure shows the effect of different levels of NPK on Banana. To achieve
this target Absolute Update Technique involves various levels, based on initial soil
test values and plant analysis (leaf nutrients).
Figure -19 shows the effect of nutrients levels of NPK on banana
plant on Nutrient Tailoring
Nutrients Tailoring
and leaf NPK
111
CHAPTER –VI
6. Comprehensive Conclusion and Scope of the Future Work
6.1 Summary of the Present Work
Banana yield prediction by The ANN based Absolute Update Technique has
been considered as the research problem in spite of the existing conventional
methods. The main reason to use ANN based model for banana yield prediction is its
model has user friendly as well as free nature. In these Experiments has tested on a
NRCB Trichy.
As it is discussed earlier, Crop yield history suggests that crop production
systems are very complex. Presently in agriculture Fertility Gradient approach is used
which speak on the analysis of past data only. It does not have any relevancy for
future prediction. Old vegetative approach gives single iteration result only. The
process of Fertility Gradient approach seems to be a much longer and complicated
process giving insufficient details to farmers in terms of accuracy in finding the
prediction of yield.
When compared with Old Vegetative methods, the ANN based Absolute
Update Technique is a modern and challenging environment method. This method is
more beneficial to the farmers and agricultural scientists for it brings in better and
accurate results in crop yield.
The Old vegetative methods obtain crop yields with at the maximum of
accuracy only, whereas the ANN based Absolute Update Technique shows
comparatively give better result and accuracy.
112
6.2 Future work
This type of research with Alluvial soil can be applied with other types of
soil as well. Similarly research can be done with other crops to test the efficiency
of Absolute Update Technique over old vegetative methods. Thereby this model
gains important not only at state level but all over India or country.
113
References
1. Adinarayana, V., D. Balagurvaiah, Y.N. Rao, P.N. Rao, I.V. Subb–Rao and
I.V.S. Rao, 1986. Potassium and sodium disorders in banana. J.Potassium Res. Reg.
Agric. Res. Sta. Lam. Guntur, Andhra Prades., 2: 102–7.
2.ALAGUMANI, T., 2005, Economic Analysis of Tissue cultured banana and Sucker
Propagated banana. Agric. Econ. Res. Rev.,, Vol. 18, January-June, pp. 81-89.
3. Ambuel, J. R., T. S. Colvin, and D. L. Karlen. 1994. A fuzzy logic yield simulator
for prescription farming. Trans. ASAE 37(6): 1999-2009.
4. Angeles, D.E., M.E. Sumner and E. Lahav, 1993. Preliminary DRIS norms for
banana. J. Plant Nutr., 16: 1059–70.
5. ANONYMOUS., 2000, Banana cultivation from tissue cultured plants. Agro India,
May 2000:24-26.
6. Annamalainagar, India, 8: 13–22 Beaufils, E.R., 1973. Diagnosis and
recommendation integrated system (DRIS). Soil Sci., University of Natal, South
Africa.
7. Arunachalam, A., N. Ramaswamy and C.R. Muthukrishnana, 1976. Studies on
the concentration in leaf tissue and fruit yield with nitrogen level.
8. Aref, S., and M. W. Wander. 1998. Long-term trends of corn yield and soil organic
matter in different crop sequences and soil fertility treatments on the Morrow Plots.
Advances in Agronomy 62: 153-197.
114
9. Barahona R. 2000. Caracterización detallada de los suelos de San Nicolás y
practicas recomendadas para su sostenibilidad. El Zamorano, Honduras.
Undergraduate Thesis. Pan-American School of Agriculture. El Zamorano.
Honduras.
10. BEGUM, J.A., AND RAHA, S.K., 2002, Marketing of banana in selected areas of
Bangladesh.Economic Affairs Calcutta, 47(3): 158-166
10. Bevege, D.I., 1978. Foliar analysis as a guide to the nutrient status of tropical
pines. In: Ferguson, A.R., R.L. Beileski and I.B. Ferguson(eds.), Plant Nutr., pp. 53–
60. Government Printer: Wellington.
12. Bongers T. 1990. The maturity index: an ecological measure of environmental
disturbance based on nematode species composition. Oecologia 83: 14-19.
13. Brady, N.C. and R.R. Weil, 2004. The Nature and Properties of Soils. 13thed.
Pearson Education, Patparganj Delhi, India.
14.Corn Hybrid Test Results in Illinois. 1967 - 1997. Report of Crop Science
Department. Urbana, Ill.: University of Illinois.
15. Cueva F. 2003. Caracterización de suelos y fertilización de frutales de altitud
media alta en el Occidente de Honduras. Undergraduate Thesis. Pan-American
School of Agriculture. El Zamorano. Honduras.
16. Delvaux B. 1995. Soils. Pp. 230-257 in Bananas and Plantains. (S. R. Gowen,
ed.). Chapman and Hall, London.
115
17. De Jong, K. A. 1975. Analysis of the behavior of a class of genetic adaptive
systems. Ph.D thesis. Department of Computer and Communication Science,
University of Michigan.
18. DHAKATE, S.M., 1996, Economic analysis of banana production in Akola
district of Maharashtra. Thesis Abstracts, 22 (1): 5-6.
19. Doran J.W. & D.C. Parkin. 1994. Defining and assessing soil quality. Pp. 6, 9 and
11 in Defining soil quality for a sustainable environment. (J.W. Doran, J.A.E. Molina
& R.F. Harris, eds). Soil Science Society of America (SSSA).Special Publication No.
35. Madison, WI.
20.Doran J.W. & T.B. Parkin. 1996. Quantitative indicators of soil quality: A
minimum data set. Pp. 25-37 in Methods forAssessing Soil Quality. (J. W. Doran &
A. J. Jones, eds). Soil Science Society of America, Inc., Madison.
21. Doran J.W. 2002. Soil health and global sustainability: translating science into
practice. Agriculture, Ecosystems andEnvironment 88:119-127.
22. Drallmeier, S. G., R. H. Teyker, and K. J. Moore. 1996. Crop Science: A
Laboratory Manual. Urbana, Ill.: Crop Science Dept., The University of Illinois at
Urbana-Champaign.
23. Drummond, S. T., K. A. Sudduth, and S. J. Birrell. 1995. Analysis and correlation
methods for spatial data. ASAE Paper No. 95-1335. St. Joseph, Mich.: ASAE.
116
24. Dumas, J. and P. Martin–Prevel, 1958. Controle de nutrition des bananeraies en
Guinee (Premiers resultats). Fruits. 13:375–386. In:Lopez, A. and J. Espinosa (eds.)
2000. Manual on the Nutrition and Fertilization of Banana. p. 54. Potash and
Phosphate Institute of Canada.
25 .Dumas, J., 1959. Etude de la du bananier pour echantillonr dans less essay set les
plantations. In: Nutrition minerale et Engrais (Ed. IFAC) Paris, Refered In: Lopez,
A. and J. Espinosa (eds.) 2000. Manual on the Nutrition and Fertilization of
Banana. pp. 54–7. Potash and Phosphate Institute of Canada.
26. Dumas, J., 1960a. Controle de nutrition de quelques bananeraies dans trios
territolres africains. Fruits, 15: 277–90. In: Lopez, A. and J.Espinosa, (eds.) Manual
on the Nutrition and Fertilization of Banana. p. 54. Potash and Phosphate Institute of
Canada.
27. Dumas, J., 1960b. Etude de la feuill du bananier pour echantillonner dans les
essis et les plantations, In: Nutrition Minerale et Engrais, IRFA,Paris. pp. 54–7. In:
Lopez, A. and J. Espinosa, (eds.) Manual on the Nutrition and Fertilization of
Banana. p. 54. Potash and Phosphate Institute of Canada.
28. Fernandez–Falcon, M. and R.L. Fox, 1985. Effect of nitrogen and potassium
fertilization on the yield of banana. Anales–de–Edafologia–Y–Agrobiologia,44:
1439–52.
29. Fernández J. 2003. Caracterización detallada de suelos de los sectores de Zorrales,
Monte Redondo de El Zamorano para el establecimiento y renovación de pasturas.
Undergraduate Thesis. Pan-American School of Agriculture. El zamorano,
Honduras.
30.Ferris H., T. Bongers & R.G.M. de Goede. 2001. A framework for soil food web
diagnostics: extension of the nematodefaunal analysis concept. Applied Soil Ecology
18:13-29.
117
31.FLORENCE WAMBUGU., 2004, Food, Nutrition and Economic Empowerment :
The Case for Scaling up the Tissue Culture Banana. Project to the Rest Africa. Paper
presented at the NEPAD/IGAD regional conference “Agricultural Successes in the
Greater Horn of Africa” Nairobi November 22-25, 2004
32.Fogain R. 2001. Nematodes and weevil of bananas and plantains in Cameroon:
occurrence, importance and host susceptibility.International Journal of Pest
Management 47:201-205.
33. Fox, R., B. King and G. Wilson, 1979. A comparative study of the sulfur nutrition
of banana and plantation. Fruits, 34: 525–34.
34.Freebairn D.M. & C.A. King. 2003. Reflections on collectively working toward
sustainability: indicators for indicators! Australian Journal of Experimental
Agriculture 43:233-238.
35. GAJANANA, T.M., 2002, Marketing practices and post-harvest loss assessment
of banana var Poovan in Tamil Nadu. Agric. Econ. Res. Rev.,, 15(1): 56-65
36. GULEDGUDDA, S.S., SHRIPAD VISHWESHWAR NAD OLEKAR, J.N.,
2002, Economics of banana cultivation and its marketing in Haveri district of
Karnataka state. Ind. J. Agric. Marketing,, 16 (1) 51-59.
37. Havlin, J.L., J.D. Beaton, S.L. Tisdale and W.L. Nelson, 2004. Soil Fertility and
Fertilizers. 6th ed. Pearson Education, Patparganj Delhi, India.
38. Haykin, S. 1994. Neural Networks: A Comprehensive Foundation. New York,
N.Y.: Macmillan College Publishing.Important factors used in deriving the fertilizer
adjustment Ramamoorthy et.el (1995) equations.
118
39. Hewitt, C.W., 1953. Leaf analysis: Bananas, Department of Agriculture Jamaica,
Bull. No. 53: 10.
40. Hewitt, C.W. and R.E. Osborne, 1962. Further field studies on leaf analysis of
"Lacatan" bananas as a guide to the nutrition of the plant. Emp. J.Exp. Agric., 30:
249–56.
41.Hollinger, S., E. D. Nafgizer, G. E. Pepper, D. W. Graffis, F. W. Simmons, R. G.
Hoeft, J. C. Siemens, M. D. McGlamery, A. G. Hager, E. L. Knake, D. R. Pike, K. L.
Steffey, M. E. Gray, and H. W. Kirby. 1997. Illinois Agronomy Handbook. Urbana,
Ill.: Crop Sciences Dept., The University of Illinois at Urbana-Champaign.
42. Jambulingam, A.R., N. Ramaswamy and C.R. Muthukrishnana, 1975.Studies on
the effect of potassium on Robusta banana. Potash Review, 27: 4–6 .
43. Jones, J.B., B. Wolf and H.A. Mills, 1991. Interpretation of results. In:
PlantAnalysis
44. Jones, J.B., B. Wolf and H.A. Mills, 1991. Interpretation of results. In: Plant
Analysis Handbook – a practical sampling, preparation, analysis,and interpretation
guide. Micro–Macro Publishing Inc., USA.
45. Jones, J.B., B. Wolf and H.A. Mills, 1991. Interpretation of results. In: Plant
Analysis Handbook – a practical sampling, preparation, analysis,and interpretation
guide. Micro–Macro Publishing Inc., USA.
46. Jones, J.B., 1997. Field sampling procedures for conducting a plant analysis.In.
Handbook of Reference Methods for Plant Analysis. Kalra, Y. P. (ed.). CRC Press,
USA.
119
47..Jones, J. B., 1998. Plant Nutrition Manual. CRC Press, USA.
48. KAMESWARA RAO, G., 2000, Comparative economics of banana and
sugarcane cultivation in Tungabhadra command area of Karnataka. M.Sc. (Agri)
Thesis, Univ.
Agric. Sci., Dharwad (India).
49. Kelling, K.A., S.M. Combs and J.B. Peters, 2000. Plant Analysis as a iagnostic
tool.http://www.soils.wisc.edu/extension/publications/horizons/2000/Plant%20Analys
is%20as%20Tool.pdf.
50.Kotur, S.C. and M.M. Mustaffa, 1984. Effect of spacing and nitrogen on the leaf
nutrient status of Robusta banana. South Ind. Hort. Central Hort. Exp., Karnataka, 32:
324–6.
51. Krantz, B.A., W.L. Nelson and L.F. Burkhart, 1948. Plant tissue tests as a tool in
agronomic research. In: Kitchen, H.B. (ed.) Diagnostic Techniques for Soils and
Crops. pp. 137–56. The Amer. Potash Inst.,Washington, D.C.
52.Lacroix, R., F. Dalehi, X. Z. Yang, and K. M. Wade. 1997. Effects of data
preprocessing on the performance of artificial neural networks for dairy yield
prediction and cow culling classification. Trans. ASAE 40(3): 839-846.
53. Lahav, E., 1970. The role of plant analysis and metabolic indicators in the
etermination of potassium levels in the banana plant. Ph.D. Thesis,Hebrew Univ.
Jerusalem.
54.Lahav, E., 1972a. Factors influencing the potassium content of the third leaf of the
banana sucker. Fruits, 27: 585–90
120
55.Lahav, E., 1972b. The role of plant analysis in determining the potassium level of
th banana plant. Fruits, 27: 855–64.
56. Lahav, E., 1977. The suitability of sampling the leaf petiole for determining the
mineral content of the banana sucker. Fruits, 32: 297–307
57. Lahav, E. and D. Turner, 1983. Banana nutrition. International Potash Institute,
Berne. Bull., 7: 1–62.
58. Lahav, E. and D. Turner, 1989. Banana nutrition. 2nd ed. International
Potash Institute. Berne. Bull., 7: 5–55.
59. Lahav, E. and D. Turner, 1992. Fertilizacion del banano para rendimentos altos.
Segunda edicion. Instituto de la Potasa y el Fosforo. Quito, Ecuador. Boletin 7:71. In:
Lopez, A. and J. Espinosa, (eds.) Manual on the Nutrition and Fertilization of
Banana. p. 49. Potash and Phosphate Institute of Canada.
60.Macy, P., 1936. The quantitative mineral nutrient requirements of plants.Plant
Physiol., 2: 749–64.
61.MADALIA, V. K., 1987, Development programme and production economics of
banana cultivation: A case study of Surat district in Gujarat state. Ind. J. Agric.
Econ.,42 (3): p 460.
62. Malavolta, E., 1994. Diagnostico foliar. In: Silva, F. (ed.) Fertilidad de Suelos:
Diagnosticoy control. Sociedad Colombia de la Ciencia del Suelo. pp. 59–98. Bogota,
Colombia.
63.MALI, B.K., BHOSALE, S.S., SHENDAGE, P.N.AND KALE, P.V., 2003,
Economics of production and marketing of banana in Jalgaon district of Western
Maharashtra, Ind. J. Agric. Econ., 17 (1):173-181.
121
64.Martin–Prevel, P., 1974. Les methodes d'echantillonnage pourl 'analyse foliaire du
bananier: resultats d'une enquete internationale et propositions envued'une reference
commune. Fruits, 29: 583–8. In:Lopez, A. and J. Espinosa, (eds.) Manual on the
Nutrition and Fertilization of Banana. p. 54. Potash and Phosphate Institute of
Canada.
65. Martin–Prevel,P.,1977. Echantillonnage du bananier pour l'analyse foliaire:
consequences des differences de techniques. Fruits, 32: 151–66. In: Lopez, A. and J.
Espinosa, (eds.) Manual on the Nutrition and Fertilization of Banana. p. 54. Potash
and Phosphate Institute of Canada.
66. Marsh S.P. 1998. What can agricultural researchers do to encourage the adoption
of sustainable farming systems.Sustainability of Farming Systems with Focus on
Future Management of Water and Nutrients, Perth, Western Australia.
67. MAURYA, O.P., SINGH, G.N. AND KUSHWAHA, R.K.S., 1996, Profitability
of banana plantation in district Hijapur (Bihar).The Bihar J. Agric. Marketing. 4
(1):68-70.
68. Memon, N., Zia-ul-Hassan and F.C. Oad, 2001. Banana nutrition management
through plant analysis. Pak. J. App.Sci. 1:563-74.
69. Memon, N.,K.S. MENON, Zia-ul-Hassan and F.C. Oad, 2005. Plant Analysis as
diagnost Tool for Evaluating Nutritional Requirements of Bananas. Pak. J. App.Sci.
5:824-831.
70.Mehrotra, K., C. K. Mohan, and S. Ranka. 1997. Elements of Artificial Neural
Networks. Cambridge, Mass.: MIT Press.
122
71.Mishra et al (2000) conducted a study on production and marketing of banana in
Gorakhpur district of Uttar Pradesh.
72. MORE, S.S.,1999, Economics of production and marketing of banana in
Maharashtra state.M. Sc. (Agri.) Thesis, Univ. Agric. Sci., Dharwad (India).
73. MORE, S.S., SHELKE, R.D.ANDKALYANKAR, S., 2005, Labour utilization
and input use pattern in banana cultivation. Agric. Marketing. 48(1): 21-23.
74. Morrow Plots. 1984. The Morrow Plots: A Century of Learning . Bulletin 775.
Urbana, Ill.: Agricultural Experiment Station, College of Agriculture, The University
of Illinois at Urbana-Champaign.
75. Orellana S. 2003. Caracterización de suelos y fertilización para frutales de clima
templado en el Occidente de Honduras. Undergraduate Thesis. Pan-American School
of Agriculture, El Zamorano, Honduras.
76. Paz, J. O., W. D. Batchelor, T. S. Colvin, S. D. Logsdon, T. C. Kaspar, and D. L.
Karlen. 1998. Analysis of water stress effects causing spatial yield variability in
soybeans. Trans. ASAE 41(5): 1527-1534.
77. Pepper, G. E. 1996. Principles of Crop Production. Urbana, Ill.: Crop Science
Dept., The University of Illinois at Urbana-Champaign.
78. QAIM, M., 1993, A socioeconomic outlook on tissue culture technology in
Kenyan banana production. Biotechnology and Development Monitor, 40:18-22.
123
78. Ramamoorthy et al.,(1967 ) reported that economic of fertilizer application was
depend on these parameter. It was on this basis that prescription method of fertilizer
recommendation for targeted yields of corn was advocated by Truog(1960).
79. RANE, A.A. AND BAGADE, S.R., 2006, Economics of production and
marketing of banana in Sindhudurg district, Maharashtra. Ind. J. Agric. Econ., 20 (1):
38-45.
80. Rashid, A., 2005. Soils: Basic concepts and principles. In: Soil Science. Memon,
K.S. and A. Rashid, (eds.). National Book Foundation,Islamabad.
81. Sarrantonio M., J.W. Doran, M.A. Liebig & J.J. Halvorson. 1996. On-farm
assessment of soil quality and health. Pp.83-105 in Methods for Assessing Soil
Quality. (J.W. Doran & A.J. Jones, eds). Soil Science Society of America,Madison.
82. Schloter M., O. Dilly & J.C. Munch. 2003. Indicators for evaluating soil quality.
Agriculture, Ecosystems and Environment 98:255-262.
83. SENTHILATHAN, S. AND SRINIVASAN, R., 1994, Production and marketing
of Poovan banana in Trichirapalli district of Tamil Nadu. Ind. J. Agric. Econ., (1): 46-
53.
84. Shivanand (2002) studied the resource use efficiency of banana crop in northern
Karnataka. He employed Cobb-Douglas type of production function, where banana
yield as dependent and land, labour, FYM, bullock labour, fertilizers plant protection
chemicals as independent variables.
85. Sikora L.J. & D.E. Stott. 1996. Soil organic carbon and nitrogen. Pp. 157-167 in
Methods of Assessing Soil Quality.(J.W. Doran & A.J. Jones, eds). Soil Science
Society of America, Madison.
124
86. SILVA,M.DE.C.A.DA., TARSITANO,M.A.A.,AND BOLIANI,A.C., 2005,
Technical and economical analysis of Apple banana tree (Musa spp.) culture, in
thenorthwest region of Sao Paulo State. Revista-Brasileira-de-Fruticultura,27(1): 139-
142.
87.STEPHEN, G.MBOGOH, FLORENCE, M.WAMBUGUA AND SAM
WAKHUSAMS., 2002,Socio-economic impact of Biotechnology Applications: Some
lessons from the pilot Tissue culture Banana production promotion project in Kenya,
pp.1997-2002.
88. SUDHARSHAN, 1998, Tissue culture banana cultivation. Agriculture and
industry survey,April/May, pp. 19-20.
89. SUDHA, M., SRINIVASA MURTHY, D. AND GAJANANA, T.M., 2005, Post
harvest handling and marketing of banana (CV Tella Chakkara Keli) in Rajahmundry
region of Andhra Pradesh. Ind. J. Agric. Marketing, 19(1): 25-37.
90. TOMAS, E.K. AND GUPTA, S. K., 1987, Economics of banana cultivation: a
case study in Kottayam district of Kerala. Ind. J. Agric. Econ., 42 (3): p. 458.
91. VENKATESH REDDY, K.G., 1982, Comparative economics of production and
marketing ofRobusta and dwarf Cavendish varieties of banana in Bangalore district.
M.Sc.(Agri) Thesis, Univ. Agric. Sci., Bangalore (India).
92. VINOD-WANJARI AND LADANIYA, M. S., (2004), Marketing of banana in
selected districts ofIndia. Tropical Agricultural Research and Extension, 7: 126-133.
93. YADAV, M.U., NAGURE, D.V., PHUKSE, K.D. AND KALALBANDI, B.M.,
2005, A Comparative study of resource productivities and resource use efficiencies of
traditional and tissue culture banana cultivation in Parbhani district of Maharashtra
state. Karnataka J. Agric. Sci., 18(3) : 735-739.
125
94. Yeates G.W. 1999. Effects of plants on nematode community structure. Annual
Review of Phytopathology 37:127-149.
95. Yeates G.W. & T. Bongers. 1999. Nematode diversity in agroecosystems.
Agriculture, Ecosystems and Environment74:113-135.
96. Yeates G.W. 2001. Diversity of soil nematodes as an indicator of sustainability of
agricultural management. Australasian Plant Pathology Society, Nematology
Workshop, Cairns, Queensland, Australia.
97. White, H., A. R. Gallant, K. Hornik, M. Stinchcombe, and J. Wooldridge. 1992.
Artificial Neural Networks: Approximation and Learning Theory. Malden, Mass.:
Blackwell Publishers.
98. Widmer T.L., N.A. Mitkowski & G.W. Abawi. 2002. Soil organic matter and
management of plant-parasitic nematodes.Journal of Nematology 34:289-295.
99. Vision 2030 Project Director, Project Directorate for Farming Systems Research
(ICAR),Modipuram, Meerut-250 110 (U.P.), India. Typeset & Printed in: Yugantar
Prakashan Pvt. Ltd.,WH-23, Mayapuri Industrial, Area, Phase-I, New Delhi.
100. Jing Liu ,C.E. Goering ,L. Tian. 2001 A Neural Network for Setting Target Corn
Yields ,published Transactions of the ASAE Vol.44(3): 705-713,American Society of
Agricultural Engineering.
LIST OF PAPERSPRESENTED/PUBLISHED
126
LIST OF PAPERS PRESENTED /PUBLISHED
International Level
1. A simulation model for Crop Yield varying Soil and different amount of
nutrients conditions using Artificial Neural Networks at International
Conference on Computer Applications (ICCA 2010), Techno Forum Research
& Development Centre in Pondicherry.
2. Efficiency Analysis of web quality using Artificial Neural Networks at
International Conference on Computer Applications (ICCA 2010), Techno
Forum Research & Development Centre in Pondicherry.
National Level
1. Efficiency Analysis on Crop Yield under varying Soil and Land management
conditions using Artificial Neural Networks at National Conference on
Research Areas in Computer Science, SRM University ,Ramapuram Campus
,Chennai.
2. A Tentative Analysis on MLR, SMLR and Back propagation Algorithm in
Artificial Neural Network for Setting a Target on Agricultural Crop Yields at
National Seminar in Annai Veilankanni's College, Saidapet, Chennai.
3. Stochastic Modeling of Expressive Speech Synthesis using a high quality
Virtual Teacher” at National Seminar in Annai Veilankanni's College,
Saidapet, Chennai.
127
SEMINARS & WORKSHOP ATTENDED
WORKSHOP
1. Microsoft Workshop about Microsoft SQL Server ICCA 2010, Techno Forum
Research & Development Centre in Pondicherry.
SEMINARS
1. State level Conference on Current Trends in Research Field of Computer
Science Shrimathi Indira Gandhi College-Trichy.
2. National level Conference on Emerging Trends of Mathematical Techniques
and their Applications in Computer Science–Shrimati Indira Gandhi College-
Trichy.
128
APPENDIX
GLOSSARY OF TERMS
TERMS DESCRIPTION
Analysis A process of interpreting data and information.
Analysis requires data input and outputs something
based on the data, experience, and previously learned
wisdom of the people involved.
Classification Classification is a system of arranging ideas or physical
objects in hierarchal and enumerative schemes.
Complete Network A complete network is a network with maximum
density: all possible lines occur.
Component A (weak) component is a maximal (weakly) connected
subnetwork.
Data Data are the smallest units of measure. The word is
technically the plural of datum but often used as a
singular. Data are the components of information.
Graph A graph is a set of vertices and a set of lines between
Density Density is the number of lines in a simple network,
expressed as a proportion of the maximum possible
129
number of lines.
Technology Technology is the set of tools both hardware (physical)
and software (algorithms, philosphical systems, or
procedures) that help us act and think better.
Threshold The threshold of a vertex is its exposure at the time of
adoption. It is equal to the proportion of its neighbours
that have adopted earlier than this vertex.
Transposed network A transposed network is a network in which the
direction of all arcs is reversed.
Actor Actor refers to a person, organization, or nation that is
involved in a social relation. Hence, an actor is a vertex
in a social network.
Adjacency Matrix An adjacency matrix is a square matrix with one row
and one column for each vertex in a network.
Adjacent Two vertices are adjacent if they are connected by a line.
Beta Test Beta test is the computer system test prior to commercial
release. Beta testing is the last stage of testing, and
normally can involve sending the product to beta test
sites outside the company for real-world exposure or
offering the product for a free trial download over the
Internet. Beta testing is often preceded by a round of
testing called alpha testing.
130
Class Class, in the context of object oriented computer
language, is the prototype for an object in an
object-oriented language; analogous to a derived
type in a procedural language. A class may also
be considered to be a set of objects which share
a common structure and behaviour. The
structure of a class is determined by the class
variables which represent the state of an object
of that class and the behaviour is given by a set
of methods associated with the class.
Class Library Class library is a term used in the object
oriented language, whcih refers to collections of
class definitions and implementations. Software
companies like Microsoft created class libraries
for reuses in programming. Class libraries and
toolkits have the reputation of being open but
too-much-assembly-required. A best of both
worlds is to deliver a useful application
composed from a toolkit where disassembly and
reassembly for evolution is supported.
Data element definition In metadata, a data element definition is a
human readable phrase or sentence associated
with a data element within a data dictionary that
describes the meaning or semantics of a data
element. Data element definitions are critical for
external users of any data system. Good
definitions can dramatically ease the process of
mapping one set of data into another data set of
131
data. This is a core feature of distributed
computing and intelligent agent development.
Data Mapping Data mapping is the process of creating data
element mappings between two distinct data
models. Data mapping is the first step in
creating a data transformation between a data
source and a destination. For example, a
company that would like to transmit and receive
purchases and invoices with other companies
might use data mapping to create data maps
from a company's data to standardized ANSI
ASC X12 messages for items such as purchase
orders and invoices.
Data Migration Data migration refers to the translation of data
between storage types, formats, or computer
systems. Data migration is necessary when an
organization decides to use a new computing
systems or database management system that is
incompatible with the current system. Data
migration is usually performed programatically
to achieve an automated migration, freeing up
human resources from tedious tasks. It is
required when organizations or individuals
change computer systems or upgrade to new
systems.
Data Modeling Data modeling is the process of structuring and
organizing data. It defines a structure for data
that is typically implemented in a database
management system and that enables (and
limits) to enter data in that structure. Data
132
modeling is often the first step in database
design and object-oriented programming as the
designers first create a conceptual model of how
data items relate to each other. Data modeling
involves a progression from conceptual model
to logical model to physical schema.
Data Processing Data processing is a computer process that
converts data into required information. The
processing is usually assumed to be automated
and running on a computer. There are many data
processing applications, such as accounting
programs that convert raw financial data into
meaninful reports for various purpose. Another
example is customer relationship management
systems (CRM) and employee relationship data
systems.
Data Scrubbing Data scubbing, also called as data cleaning, is
the process of detecting and removing and/or
correcting a database to increase data accuracy,
reduce redundancy and enhance consistency of
different sets of data that have been merged
from separate databases. Sophisticated software
applications are available to clean a database
data using algorithms, rules and look-up tables,
a task that was once done manually and
therefore still subject to human error.
Data Structure Data structure is the pattern to store data in a
computer so that it can be used efficiently.
Often a carefully chosen data structure will
allow a more efficient algorithm to be used. The
133
choice of the data structure often begins from
the choice of an abstract data structure. A well-
designed data structure allows a variety of
critical operations to be performed, using as few
resources, both execution time and memory
space, as possible. Data structures are
implemented using the data types, references
and operations on them provided by a
programming language.
Data Transformation Data transformation converts data from a source
data format into destination data. Data
transformation can be divided into two steps: 1)
data mapping maps data elements from the
source to the destination and captures any
transformation that must occur; 2) code
generation that creates the actual transformation
program.
Database Administration Database administration refers to duties,
typically performed by a DBA in an
organization, such as disaster recovery (backups
and testing of backups), performance analysis
and tuning, and some database design or
assistance thereof.
Database Model A database model is a theory or algorithm
describing how a database is structured and
used. Several such models have been suggested,
for example, Hierarchical model, Network
model, Relational model, Object-Relational
model, Object model, Associative, Concept-
oriented, Entity-Attribute-Value, Multi-
134
dimensional model, Semi-structured, and Star
schema.
Database Normalization Databases normalization is a process that
eliminates redundancy, organizes data
efficiently, reduces the potential for anomalies
during data operations and improves data
consistency. The formal classifications used for
quantifying "how normalized" a relational
database is are called normal forms. A non-
normalized database is vulnerable to data
anomalies because it stores data redundantly. If
data is stored in two locations, but later is
updated in only one of the locations, then the
data is inconsistent; this is referred to as an
"update anomaly". A normalized database stores
non-primary key data in only one location.
Database Object Database Object is a piece of information or
record that is stored in a database.
Database Query Language Database query language is a kind of
programming language to retrieve information
from a database. The person formulating the
query is expected to understand the relevant
rules for formulating the query, and to program
the query according to the requirements.
Examples of the database query language are
the CODASYL database language, "network"
databases, relational algebra, relational calculus,
Datalog, SQL3, QUEL, XPointer, XPath and
OQL.
135
Database Server A database server is a computer program that
provides database services to other computer
programs or computers, as defined by the client-
server model. The term may also refer to a
computer dedicated to running such a program.
Database management systems frequently
provide database server functionality, and some
DBMS's (e.g., MySQL) rely exclusively on the
client-server model for database access.
Glueware Glueware is a type of software that can be used
to "glue" or integrate systems, software
components and databases together, to form a
seamless integrated system.
Gmail Drive Gmail Drive, a free shell namespace extension
("add-on") for Microsoft Windows Explorer,
makes it possible to create a new network share
on the workstation. In order to use this add-on,
you need a Gmail account from Google Gmail.
The add-on enables you to use the normal
Windows desktop file copy and paste
commands to transfer files to and from your
Gmail account just as if it was physically
located on your local network.
Generalized Markup Language Generalized Markup Language (GML) is a set
of macros (tags) for the IBM text formatter,
"SCRIPT". SCRIPT is the main component of
IBM's Document Composition Facility (DCF).
GML simplifies the description of a document
in terms of its format, organization structure and
content parts and their relationship, and other
136
properties. GML markup (or tags) describe such
parts as chapters, important sections, and less
important sections (by specifying heading
levels), paragraphs, lists, tables, and so forth.
Using GML, a document is marked up with tags
that define what the text is, in terms of
paragraphs, headers, lists, tables, and so forth.
The document can then be automatically
formatted for various devices simply by
specifying a profile for the device. For example,
it is possible to format a document for a laser
printer or a line (dot matrix) printer or for a
screen simply by specifying a profile for the
device without changing the document itself.
Handwriting Recognition Handwriting recognition refers to a computer
receiving handwritten input and intelligently
recognize it to some characters. The image of
the written text may be sensed "off line" from a
piece of paper by optical scanning (optical
character recognition). Alternatively, the
movements of the pen tip may be sensed "on
line", for example, by a pen-based computer
screen surface.
Haskell Programming Language Haskell Programming Language, simply called
Haskell in most cases, is a standardized pure
functional programming language with non-
strict semantics, named after the logician
Haskell Curry. Haskell was designed by a
committee from the functional programming
community in April 1990. It features static
137
polymorphic typing, higher-order functions,
user-defined algebraic data types, and pattern-
matching list comprehensions. Innovations
include a class system, systematic operator
overloading, a functional I/O system, functional
arrays, and separate compilation.
Helper Applications Helper application is an external viewer
program launched to display content retrieved
using a web browser. These applications
commonly let you see and hear video and audio
files, as well as view specialized text files or
virtual reality models. Windows Media Player,
QuickTime, Shockwave, CosmoPlayer, and
RealAudio are examples of helper applications.
Another common term for these programs is
"plug ins," because they supplement the
capabilities of your browser, and only run when
they are needed to display files.
Heterogeneous System Heterogeneous systems, in software context,
refer to systems that have different aspects such
as the interface, the implementation, the data,
etc. Two systems in a family are heterogeneous
to the extent that they are incompatible in some
way. One may represent information differently
or not include certain functionality or adopt
different security policies. If everything between
two systems are the same and interoperate, they
are homogeneous. Federating or integrating
homogeneous systems is presumably simpler
than federating heterogeneous systems.
138
Hex editor Hex editor is a tool used to create and modify
binary files.
High-Level Language High-level language, also known as high-level
programming language, is a programming
language that, in comparison to low-level
programming languages, may be more abstract,
easier to use, or more portable across platforms.
Examples include languages such as C,
FORTRAN, or Pascal that enables a
programmer to write programs that are more or
less independent of a particular type of
computer.
Intelligent Device Management Intelligent Device Management is a term used
for enterprise software applications that allow
various equipment manufacturers to proactively
monitor and manage remote equipment, systems
and products via the Internet and provide instant
and cost-effective service & support to their
customers.
IntelliJ IDEA IntelliJ IDEA is a commercial Java IDE
by JetBrains company. It includes a set of
integrated refactoring tools that allow
programmers to quickly redesign their code. A
number of its features accelerate development
and allow programmers to concentrate on
functionality while IntelliJ IDEA handles more
mundane coding tasks. Among other features,
IntelliJ IDEA provides close integration with
popular open source development tools such as
CVS, Subversion, Apache Ant and JUnit.
139
Interchangeability In computer science, interchangeability is an
ability that an object can be replaced by another
object without affecting code using the object.
That chance usually requires two objects share
an interface that is either same strictly or
compatible in particular case.Â
Interface An interface, in computer programming,
is a defined means for a system to communicate
with other systems. It is a boundary between a
system and its environment providing ways of
providing the system inputs and receiving
outputs. In Object Oriented programming, class
definitions and method signatures provide
interfaces. Application program interfaces
(APIs) form the interface of a system to
applications and often consist of collections of
functions or commands in a scripting language.
Interfaces may be hidden (available only to the
system developer) or exposed (available to
others).
Interface Encapsulation An interface encapsulates refers to an
implementation in a system in which the system
implementation can be changed without
changing the interface. With the interface
encapsulation property, the changes in the
system will not effect its way to communicate
with other systems.
Interface Standard Interface standard refers to a standard in
communications that defines one or more
functional and/or physical characteristics
140
necessary to allow the exchange of information
between two or more systems or equipments.
An interface standard may include operational
specifications and acceptable levels of
performance.
Information Technology Management Information technology
management (IT management), also called
Management of Information Systems (MIS), is a
combination of two branches: information
technology and management. One implies the
management of a collection of systems,
infrastructure, and information that resides in
them. Another implies the management of
information technologies as a business function.
This aims at achieving the goals and objectives
of an organisation through computers.
Information Technology Information Technology (IT) is a broad
subject concerned with technology and other
aspects of managing and processing
information, especially in large organizations. In
particular, IT deals with the use of electronic
computers and computer software to convert,
store, protect, process, transmit, and retrieve
information. For that reason, computer
professionals are often called IT specialists or
Business Process Consultants, and the division
of a company or university that deals with
software technology is often called the IT
department. Other names for the latter are
information services (IS) or management
141
information services (MIS), managed service
providers (MSP).
Java 2 Platform, Enterprise Edition Java 2 Platform, Enterprise Edition
(J2EE), now called Java Platform, Enterprise
Editor(Java EE), is a programming
platform—part of the Java Platform—for
developing and running distributed multitier
architecture Java applications, based largely on
modular software components running on an
application server. The Java EE platform is
defined by a specification. Similar to other Java
Community Process specifications, Java EE is
also considered informally to be a standard
because providers must agree to certain
conformance requirements in order to declare
their products as Java EE compliant; albeit with
no ISO or ECMA standard.
JACK Audio Connection Kit The JACK Audio Connection Kit
(JACK) is a soundserver or daemon that
provides low latency connections between so-
called jackified applications. It is created by
Paul Davis and others and licensed under the
GPL. JACK is free audio software. It can use
ALSA, PortAudio and (still experimental) OSS
as its back-end. As of 2003 it runs on GNU /
Linux and Mac OS X.
JADE Programming Language JADE is an object-oriented programming
language that exhibits a seamlessly integrated
object-oriented database management system. It
is designed to be an end-to-end development
142
environment, which allows systems to be coded
in one language from the database server at one
end down to the clients at the other.
Object Database Management System Object Database Management System
(ODBMS), also known as Object Oriented
Database Management System (OODBMS),
refers to the database management system for an
object database. Benchmarks between ODBMSs
and relational DBMSs have shown that ODBMS
can be clearly superior for certain kinds of tasks.
The main reason for this is that many operations
are performed using navigational rather than
declarative interfaces, and navigational access to
data is usually implemented very efficiently by
following pointers. Critics of ODBMS, suggest
that pointer-based techniques are optimized for
very specific "search routes" or viewpoints.
However, for general-purpose queries on the
same information, pointer-based techniques will
tend to be slower and more difficult to formulate
than relational.
Object Desktop Network The Object Desktop Network (OD or ODNT) is
a software subscription service created by
Stardock. Launched in 1995 on OS/2, it
transitioned in 1997/98 to the Windows
platform. Subscribers typically download Object
Desktop components across the Internet using
Stardock Central, although CD snapshots are
available on request. Once downloaded, users
143
may use released versions of components
forever.
Open Knowledge Initiative The Open Knowledge Initiative (O.K.I.) is an
organization responsible for the specification of
software interfaces comprising a Service
Oriented Architecture (SOA) based on high
level service definitions. The Open Knowledge
Initiate was initially sponsored by the Andrew
W. Mellon Foundation, Massachusetts Institute
of Technology and the IMS Global Learning
Consortium. O.K.I. has designed and published
a suite of software interfaces known as Open
Service Interface Definitions (OSIDs), each of
which describes a logical computing service.
Online Analytical Processing Online Analytical Processing is a type of
software that allows for the real-time analysis of
data stored in a database. It is an approach to
quickly provide the answer to analytical queries
that are dimensional in nature. The OLAP server
is normally a separate component that contains
specialized algorithms and indexing tools to
efficiently process data mining tasks with
minimal impact on database performance. The
typical applications of OLAP are in business
reporting for sales, marketing, management
reporting, business performance management
(BPM), budgeting and forecasting, financial
reporting and similar areas.
Object Linking and Embedding Object Linking and Embedding (OLE), a
technology developed by Microsoft, enables the
144
creation of documents by incorporating
elements created using different kinds of
software. Object Linking and Embedding
system allow objects from one application to be
embedded within another (eg, taking an Excel
spreadsheet and putting it into a Word
document).
Online Transaction Processing Online Transaction Processing (or OLTP) is a
class of program that facilitates and manages
transaction-oriented applications, typically for
data entry and retrieval transaction processing.
OLTP also refers to computer processing in
which the computer responds immediately to
users' requests. An automatic teller machine for
a bank is an example of transaction processing.
Probably the most widely installed OLTP
product is IBM's CICS (Customer Information
Control System).
Object Management Group Object Management Group (OMG) is a
consortium, originally aimed at setting standards
for distributed object-oriented systems, and now
focused on modeling (programs, systems and
business processes) as well as model-based
standards in some 20 vertical markets. Founded
in 1989 by eleven companies (including
Hewlett-Packard Company, Apple Computer,
American Airlines and Data General), OMG
mobilized to create a cross-compatible
distributed object standard. The goal was a
common portable and interoperable object
145
model with methods and data that work using all
types of development environments on all types
of platforms. At its founding, OMG set out to
create the initial Common Object Request
Broker Architecture (CORBA) standard which
appeared in 1991.
OmniPage Professional OmniPage Professional is software used in
conjunction with a scanner, to scan pictures or
documents into the computer.
Object-Oriented Language Object-oriented language (OO language) is a
type of computer programming language that
allows or encourages, to some degree, object-
oriented programming methods. OO languages
can be grouped into several broad classes,
determined by the extent to which they support
all features and functionality of object-
orientation and objects: classes, methods,
polymorphism, inheritance, and reusability.
Object Oriented Database Management System Object Oriented Database
Management System (OODBMS), also known
as Object Database Management System
(ODBMS), refers to the database management
system for an object database. Benchmarks
between ODBMSs and relational DBMSs have
shown that ODBMS can be clearly superior for
certain kinds of tasks. The main reason for this
is that many operations are performed using
navigational rather than declarative interfaces,
and navigational access to data is usually
implemented very efficiently by following
146
pointers. Critics of ODBMS, suggest that
pointer-based techniques are optimized for very
specific "search routes" or viewpoints.
However, for general-purpose queries on the
same information, pointer-based techniques will
tend to be slower and more difficult to formulate
than relational ones.
Object-oriented programming OOP is a computer programming paradigm, in
which writing programs in one of a class of
programming languages and techniques based
on the concept of an "object" which is a data
structure (abstract data type) encapsulated with
a set of routines, called "methods" which
operate on the data. Operations on the data can
only be performed via these methods, which are
common to all objects which are instances of a
particular "class". Thus the interface to objects
is well defined, and allows the code
implementing the methods to be changed so
long as the interface remains the same. The
programming languages support object-oriented
programming, including the Java platform and
the .NET Framework.