Mapping Rice Fallow Areas for Short Season Grain Legumes Intensification in South Asia using MODIS 250m Time-Series Data

Include Partner Logos, if any

Leveraging legumes to combat poverty, hunger, malnutrition and environmental degradation

Gumma MK1,Thenkbaail PS2, Teluguntla P2, Xiong J2, Mohammed I1 and Whitbread A1

MATERIALS AND METHODS - Datasets: Satellite imagery: MODIS-250m, 16-day reflectance composites were used to derive Normalized Difference Vegetation Index (NDVI), Land Surface Water Index (LSWI) and the NDVI Monthly Maximum Value Composites (NDVI-MVC) for the year 2014-15. Extensive ground survey information: Ground information was collected during different seasons for two distinct purposes, on land use/land cover including irrigation source, crop intensity and crop extent. The first set of field points (996 (132 +361) locations) were collected during October 11-20, 2010 for mapping cropland areas and second set of ground dataset (402 (131 +467) locations) were collected during January 15-30, 2014 for validation. Methods: The NDVI plots are ideal for understanding the changes within and between cropping seasons and between classes and exhibits the length of growing period. Temporal NDVI signature clearly elicits the planting time, peak growth and harvesting stage in Figure 1. Detailed methods and approaches were explained in Gumma et al 2016

DISCUSSIONS– Discuss the results, its achievements, include challenges, learnings and way forward. (Font: Calibri; 36 pts. approx. 200 words)

REFERENCES – Gumma, M.K., Thenkabail, P.S., Teluguntla, P., Rao, M.N., Mohammed, I.A., Whitbread, A.M., 2016. Mapping rice-fallow cropland areas for

short-season grain legumes intensification in South Asia using MODIS 250 m time-series data. International Journal of Digital Earth 9(10), 981-1003.

• RESULTS - Rice fallows were mapped in South Asia using NDVI based temporal profiles, phenological signatures (Fig. 1) • Developed a land use map (Fig. 2) for 2014-15. Almost all of the intermixing or misclassification was between various

rainfed rice classes.

ABSTRACT:The goal of this study was to map rainfed and irrigated rice-fallow cropland areas across South Asia using MODIS 250m time-series for intensifying short season grain legumes such as chickpea. The suitability of rainfed and irrigated rice-fallow croplands for grain legume cultivation across South Asia were defined as areas that were identified as: (a) rice during the monsoon season, (b) fallow during the rabi season, and (c) Low water-consuming grain legumes during rabi season. Existing irrigated or rainfed crops other than rice that were grown during kharif were not considered suitable for grain legumes within the context of this study, because the moisture demand of these crops is too high for the rabi season. The study analyzed moderate-resolution imaging spectroradiometer (MODIS) 16-day 250 meter normalized difference vegetation index time-series data acquired for the period June 2010 to May 2011 using spectral matching techniques. Accuracy was evaluated based on independent ground survey data and compared with available sub-national level statistics. The accuracies of the cropland fallow classes were between 75-82%. The analysis estimated approximately 22.3 million ha of suitable rice fallow areas in South Asia. These areas can be targeted by decision-makers for sustainable intensification of short duration grain legumes.

INTRODUCTION - Cereal grains (e.g., rice, wheat) are most extensively cultivated crops in South Asia. Vast extent of these croplands where grain legumes are grown during Kharif (June-October) season, are left fallow during the Rabi (November-February), and\or the Summer (March-May) due to insufficient water during these seasons to grow water demanding cereal crops. However, far less water demanding grain legumes (e.g., Chickpea) can be grown during Rabi and\or summer seasons in these fallows. Such a measure will help farmers supplement their income, increase food security, and enrich diverse nutritional demand of increasingly richer South Asian population.

1 International Crops Research Institute for the Semi-Arid Tropics, Patancheru, India;2 U.S. Geological Survey (USGS), Western Geographic Science Center, Flagstaff, AZ, USA.

Cropland distribution 2014-15 (Percent of total cropland (%))

01. Irrigated-SC-rice in kharif-fallow in rabi-fallow in summer (4)

02. Irrigated-SC-fallow in kharif-rice in rabi-fallow in summer (1)

03. Irrigated-DC-rice in kharif-mixed crops in rabi-fallows in summer (10)

04. Irrigated-DC-rice in kharif-ric in rabi-fallow in summer (1)

05. Irrigated-TC-rice in kharif-mixed crops in rabi-rice in summer (4)

06. Rainfed-SC-rice in kharif-fallow in rabi-fallow in summer (6)

07. Rainfed-SC-fallow in kharif-rice in rabi-fallow in summer (0.4)

08. Rainfed-SC-flooded in kharif-flooded in rabi floded-summer rice (0.3)

09. Irrigated-DC-mixed crops in kharif-mixed crops in rabi-fallow in summer (35)

10. Rainfed-SC-mixed crops in kharif-fallow in rabi-fallow in summer (38)

11.Other LULC

))(exp(1))(exp(1 qtd

vvk

ptc

kvNDVI es

st

(1)

Fig.1: Classic case of rainfed rabi fallow. A model of vegetation phenology and transition dates, as in equation (1). Tmin defines the beginning of the time series, Ton is onset of greenness, Tdev beginning of development stage, Tsen onset of senescence, and Thar is harvesting of senescence. p and q are the inflection points. Figure shows the progression of class 06. Rainfed, single crop during kharif (June-October), fallow during rabi (November-February), and fallow during summer (March-May) (Figure 2). In this study we want to map rice croplands areas (either in kharif or summer) that are left fallow during rabi.

Fig.2: Spatial distribution of rice-systems and rice-fallows (rabi-fallows)

For more information: m.gumma@cgiar.org

Country Area in (000' ha)

01. Irrigated-rice-fallows

06. Rainfed-rice-fallows

Total

Bangladesh 1190 739 1929 Bhutan 3 1 4 India 7909 11695 19604 Nepal 194 117 311 Pakistan 14 93 107 Sri Lanka 244 1 245 Total 9554 12646 22200