Aridity Argentina

8/3/2019 Aridity Argentina

1/25

Aridity and desertification: Exploring environmentalhazards in Jachal, Argentina

Susana B. Adamo a,*, Kelley A. Crews-Meyer b

a Carolina Population Center, University of North Carolina, 302E University Square East,

123 West Franklin Street, Chapel Hill, NC 27516, USAb Department of Geography and the Environment, University of Texas, 210 W.

24th #334, MC A3100, Austin, TX 78712, USA

Abstract

This paper explores environmental hazards, more specifically desertification processes, in an area of west

central Argentina, addressing the combined influence of the physical framework and the long lasting human

settlement and use of natural resources. It is based upon the analysis of remotely sensed using vegetation

indices, image differentiation, change detection, and pattern metrics. The results indicate a net decreased in

the amount of vegetation between 1973 and 2001, and increasing fragmentation of vegetation classes. Thisis interpreted as a sign of the presence of land degradation processes likely linked to human activities in the

areas of irrigated farming, grazing, firewood gathering and population settlement.

q 2005 Elsevier Ltd. All rights reserved.

Keywords: Arid lands; Desertification; Aridity; Argentina; Environmental hazards; Remote sensing; Natural resources

management

Introduction

Desertification represents one of the most threatening environmental hazards due to the largeamount of people and land at risk1. Although numbers vary across the literature, it is estimated

that desertification directly affects one-third of the Earths land and 250 million people, most of

them in developing countries; even a higher number could be potentially affected (Clarke &

Noin, 1998; Murray, Burke, Tunstall, & Gilruth, 1999; UNCCD, 2004).

Applied Geography 26 (2006) 6185

www.elsevier.com/locate/apgeog

0143-6228/$ - see front matter q 2005 Elsevier Ltd. All rights reserved.

doi:10.1016/j.apgeog.2005.09.001

* Corresponding author. Tel.:C1 919 966 6835; fax: C1 919 962 7217.

E-mail address: [email protected] (S. B. Adamo).1 Defined in a broad sense, environmental hazards include all those physical agents, events, objective conditions, and

processes in the environment that create threats to livelihoods or lives, regardless of their genesis (Hewitt, 1997:25). The

concept takes account of processes originated from natural resource management issues (e.g. deforestation) as well as ofpotentially dangerous natural ecosystem dynamics (e.g. earthquakes).
http://www.elsevier.com/locate/apgeoghttp://www.elsevier.com/locate/apgeog


2/25

Desertification is the label for land degradation in arid, semiarid, and dry sub-humid areas,

collectively called drylands. Land degradation is defined as the reduction or loss of the

biological or economic productivity and complexity of agricultural land (including rainfed

cropland, irrigated cropland, range and pasture), forests and woodlands (United Nations,

1994:4-5). The term implies a negative change with respect to previous environmentalconditions, often implicitly considered as the normal or desirable state of affairs in the

ecosystem in terms of its use and management (Blaikie & Brookfield, 1987).

Desertification is the outcome of different processes, including climatic variation and human

activities (United Nations, 1994:4). Regarding climatic variation, moisture deficiency or aridity

is the most common climatic criteria used for the definition and classification of drylands, and is

caused when water losses (evaporation and transpiration, referred collectively as evapotran-

spiration) exceed water inputs (precipitation) (Strahler & Strahler, 1992:164). Low and irregular

rainfall, high evapotranspiration, and cyclical droughts contribute to this deficiency.2 Human

activities leading to land degradation in drylands are related to land uses and habitation patterns,and include: (a) soil erosion caused by wind and/or water; (b) deterioration of the physical,

chemical and biological, or economic properties of soil; and (c) long-term loss of natural

vegetation (United Nations, 1994:5).

An immediate consequence of the United Nations definition of desertification is that its

understanding requires taking into account two different but related aspects: (1) the natural

dynamics, average characteristics, and variations of the physical and biological components of

arid and semiarid environments, their evolution over time, and their spatial diversity; and (2)

how populations have historically used the natural resources in a given area, and what kind of

environmental problems may stem from different forms of land use and management.

A caveat in considering desertification is that, because the ecological and social dynamics andvariability of drylands are not that well known, the determination of what states of the ecosystem

should be regarded as problematic in terms of productivity is not an easy task (Agnew, 2002; Dregne,

2002; Thomas & Middleton, 1994). The carrying capacity of drylands is generally low because of the

reduced availability of water derived from aridity, and this scarcity in turn affects the availability, use,

and management of other natural capital components, e.g. land and biomass (Knerr, 1998:213). In

addition, desertification is a process characterized by cumulative changes or slow-onset disruption

(International Organization for Migration, 1992:12), making it difficult to identify its initial stages.

In this paper, we seek to contribute to the knowledge of the environmental and social

dynamics of drylands, by exploring environmental hazards related to aridity and desertification

in the department of Jachal, an arid area located in the province of San Juan in the west of

Argentina3. The general objective is twofold. We intend to address environmental change,

identifying and locating land degradation processes from 1973 to 2001. Also, we intend to

establish the relation between these processes, the natural dynamics of the ecosystem, and

human activities and settlement.

2 Defining aridity involves the calculation of the ratio of precipitation to potential evapotranspiration (P/PET), also

called an aridity index, which should lie between 0.05 and 0.65 for an area to be considered dryland. There are different

ways of deriving this index from climatic data, but the underlying principles are basically the same (Middleton &

Thomas, 1997:2; Noin & Clarke, 1998:2; UNEP, 1992:2; United Nations, 1994:5; UNSO, 1997:5). Other definitions ofdrylands take into account only the amount of annual rainfall. The limit between sub-humid and humid areas is usually

set at 800 mm (Little, 1994).3 In Argentina, departments are third level administrative units, after country and province, similar to counties in the

US.

S. B. Adamo, K. A. Crews-Meyer / Applied Geography 26 (2006) 618562


3/25

The structure of the paper is as follows. After a brief overview of arid lands in Argentina, we

introduce the study area, including environmental conditions and dynamics, natural resource

management, and documented environmental problems in the department. This will set a

baseline or starting point for discussing environmental change. In Section 3, we describe the data

and methods used, followed by the presentation of the results of the remote sensing analyses.Sections 5 and 6 offers the discussion and conclusions of the study.

Aridity and desertification in Argentina

Although Argentinas image is generally linked to the green prairies of the Pampas, it actually

may be defined as an arid country. An estimated 70% of its area is in arid, semiarid, or dry-

subhumid lands, where only 12% of the countrys water resources are located. Almost 30% of its

population lived in those areas in 1991, and drylands represent 50% of the agricultural value

(Abraham, 2003; Deng, 2000; Dregne & Chou, 1992; Murray et al., 1999). A substantial amountof the land in Argentinas drylands suffers erosive process classified as moderate or severe,

resulting in declining agriculture productivity and deteriorating living conditions (Abraham,

2003). The population in areas under desertification amounts to almost seven million people or

19% of the countrys population, and poverty and indigence in these areas are consistently

higher than the national averages (CEPAL/GTZ, 2003).

In the province of San Juan, where the study area is located, environmental degradation is

considered significant. It is attributed to a combination of both environmental characteristics (i.e.

arid climate and highly varied topography) and natural resources management issues (notably

water management, land ownership and lack of modern technology) (PROSA, 1988; SAGyP &

CFA, 1995).

The study area

The department of Jachal is located in the north of the province of San Juan, in the heart of

Argentinas drylands4. Its area is 14,749 km2 (5,590 square miles), and it had a total population

of 20,898 in 20015. The proportion of rural population was 48%, remarkably higher that the

national percentage (11%). The only urban centre is San Jose de Jachal, founded in 1751 and a

small city of 10,901 inhabitants in 2001. The rest of the population is distributed in small rural

towns and villages in the main and secondary irrigation areas, along the rivers. Very few people

live dispersed in the countryside, along the main roads, and in the puestos, temporary grazing

settlements in the Precordillera natural pastures. The locations of the population centers and

puestos are displayed in Fig. 1.

The environmental framework

Located within the Arid Diagonal of South America, aridity is the defining characteristic of

Jachals ecosystem, and the main ecological limiting factors are related to water availability. The

average total annual precipitation oscillated between 100 and 200 mm for the 19411990 series,

4 Jachal is the name of the department (administrative unit), the main river, the only urban area, and the main

irrigation district.5 This is the date of the last population census.

S. B. Adamo, K. A. Crews-Meyer / Applied Geography 26 (2006) 6185 63


4/25

which places the department well into the definition of an arid area6. The evapotranspiration for

the same period averaged more than 1000 mm (Cornejo, 1997), yielding an aridity index (ratio

of precipitation to potential evapotranspiration) of approximately 0.09. This score not only

classifies this area as arid, but even borders on hyper-arid according to the UNSO classification.In addition to this, precipitation distribution along the year is irregular, concentrated in a few

months or even weeks during the summer (JanuaryMarch). Wide interannual variations in

rainfall are common and there are also spatial variations due to differences in altitude,

geomorphology, and relief orientation across the large area of the department. High

evapotranspiration is stimulated by the hot summers, temperate winters and wide daily

temperature fluctuations that also characterize the climate of Jachal. Most of the year is frost

free, representing a potential advantage in the growth and marketing of vegetables such as

onions and tomatoes (Allub & Guzman, 2000). The constant water deficit makes agriculture

impossible without irrigation, and also poses a heavy burden on the population in terms of water

Fig. 1. The Department of Jachal. The geomorphology of Jachal is complex, including the Precordillera on the west, the

Sierras Pampeanas on the east, their foothills and piedmonts, the intermountain valley where the irrigation districts of

Jachal, Niquivil and Tucunuco are located, and the plains of the east that form the desolated T ravesa of Ampacama.

Source: Authors elaboration, after Lizana (2001).

6 Using the precipitation criterion, drylands include the following zones: (a) sub-humid, 5001000 mm, (b) semiarid,

250500 mm, and (c) arid, 0250 mm (Strahler & Strahler, 1992:146).



5/25

availability for household needs such as drinking and cleaning. However, the small amount of

local precipitation is sufficient to maintain the native vegetation, which represents both a forage

resource for small goat herds and firewood for the human population.

Water availability is tied to surface watercourses, and permanent rivers are scarce. The largest

perennial stream is the Jachal river, a typical allochthonous7 river (see location in Fig. 1).

Average monthly discharge varies between 8 and 9 m3 /sec (cm) (Allub & Guzman, 2000;

Pannocchia, 1979), quite low for being the areas most reliable water source. There are strong

seasonal variations in discharge, as can be observed in Fig. 2. Summer months represent the peak

discharge season while winter months are the low discharge period, a pattern derived from the

rivers source, snowmelt in the Arid Andes. The river discharge also presents strong interannual

variations. While average discharge for January is 15 cm, in January of 1942 the river discharged144 cm and in January of 1971, 3.40 cm. A particular problem related to the Jachal river is its

high salinity (particularly sodium and boron), which adds to the ecological constraints, reducing

yields and limiting viable crops to those tolerant to boron (onions, alfalfa, wheat, corn and

olives).8

After water availability, soils are the most crucial natural resource for agriculture. In Jachal, soil

quality is not optimal due to the climatic constraints: soil formation is slower in arid environments

because of water and organic matter scarcity, and the resultant soils are generally thin and of low

010

20

30

40

50

60

70

80

90

100

110

120

jan

apr

jul

oct

jan

apr

jul

oct

jan

apr

jul

oct

jan

apr

jul

oct

jan

apr

jul

oct

months

cms

1940-44 1967-72 Average

Fig. 2. Jachal river: average monthly discharge, and discharge in humid (19401944) and dry (19671972) years. Source:

Calculations based on data from the Direccion Provincial de Hidraulica of San Juan. The measure point was the stream

gauging station of Pachimoco, near San Jose de Jachal. Note: CMS stands for cubic meter per second.

7

Allochthonous applies to watercourses whose source and upper reaches are located in a different and generallymoister environment than the one where they end, and which receive no or very few tributaries when they pass through

arid lands, being greatly reduced by evaporation and often by the use of their waters for irrigation (Mainguet, 1999:85).8 Boron concentration reaches 4.5 mg/litre when the maximum limit has been set at 3.75 mg/litre (Centro Regional del

Agua Subterranea, 1972).



6/25

agricultural potential (Mainguet, 1999:46). Fig. 3 displays soil distribution in Jachal. A substantial

amount of the department area is categorized as rock, with little to no soil layer and consequently

not suitable for agriculture purposes. In the rest of the department, soils have been classified within

the order Entisols, incomplete soils due to the physical characteristics of the environment:

topography, erosion processes, aridity, and scarce organic matter (INTA, 1990; INTA/Aeroterra,

1995; Strahler & Strahler, 1992:46362, 58182). Because these characteristics and processes vary

across Jachals area, three groups of Entisols are found: torrifluvents, torriorthents and

torripsaments. They exhibit variations in their capacity to sustain agricultural activities, always

within the overall low productivity that characterizes dryland soils. These variations are related todifferences in chemical composition, to micro-level combinations of climate and geomorphology,

and in some cases to decades of irrigated agriculture.

As field-based measurements of soil capacity for sustaining agriculture do not exist for this

area, an indirect measurement is the productivity index or PI. This indicator is interpreted as a

proportion of the highest potential yield or productivity of typical crops in the area. In other

words, (100-PI) is the percentage of decline in the optimal yield due to some characteristic or

limitation (INTA, 1990; INTA/Aeroterra, 1995).9 PIs are generally low in the department,

Fig. 3. Soils of the Department of Jachal. Soils in Jachal have been classified into three groups of the order Entisols.

Torrifluvents are soils of arid climate that are not flooded frequently or for a long time. Torriorthents, located in the

piedmonts surrounding the central oasis, include dry or salty soils on moderate or high slopes. Torripsaments present a

sandy texture and low water retention capacity (INTA/Aeroterra, 1995; Strahler & Strahler, 1992). Source: Authors

elaboration after INTA (1990) and INTA/Aeroterra (1995).

9 The productivity index PI is calculated taking in account climatic factors (such as temperature and precipitation) (H),

and soil factors (such as drainage D, depth P, texture T, salinity Sa, organic matter M, erosion E, etc.). These factors are

included in a multiplicative parameter: PIZH!D!P!T. and so on. The climatic factor (H) for San Juan was set in 65

due to aridity, which in turn resulted in PIs below 35 for the provinces soils ( INTA, 1990).



7/25

but the index presents wide differences by type of soil and location. Torrifluvents, torripsaments

and torriorthents show PIs of 33, 18 and 4, respectively. Within these averages, soils located in

the western or mountainous portion of the study area have lower PIs regardless of type, due to

the effect of slopes and gravity on soil formation.

Influenced by water deficiencies and variability, and soil conditions, the natural vegetationcover is a combination of sparse open woods, shrub steppes, and wide patches of bare soil,

arrange in an irregular spatial pattern that follows micro-level changes in ecological conditions

(Cabrera, 1958). It corresponds to the floristic Province of Monte, and displays a combination of

open woods and shrub steppes identified respectively by two species, algarrobo ( Prosopis spp.)

and jarilla (Larrea spp.). Species are xerophytes well adapted to aridity, and some of them are

adapted to live in particular niches within the general arid environment (rocks, sand, clay, salt

and marshes) (Cabrera, 1958; Morello, 1958). Halophytes (salt tolerant species) like jume

(Allenrolfea vaginata) and zampa (Atriplex lampa) could be considered biological markers for

land degradation, since they usually replace xerophytes in areas of waterlogging and acute

salinization (Centro Regional de Agua Subterranea, 1972; Kiesling, 1994).

The use and (Mis)management of natural resources

Water, land, and vegetation cover represent the natural resource base of Jachals economy

and households livelihoods and consumption. Over time, different environmental problems

have been linked to the use and management of these natural resources through, for example,

irrigated farming and ranching, grazing in natural pastures, and gathering of firewood for heating

and cooking. A number of processes in Jachal could eventually be linked to desertification or

land degradation, among them salinization of soils and water, waterlogging, and deforestation(Allub & Guzman, 2000; Pannocchia, 1979). It has been argued that this degradation has had

negative economic effects for the department (PROSA, 1988), and local authorities are now

involved in the recuperation of soils affected by salinization, waterlogging and nutrient depletion

(Davire de Musri & Malberti de Lopez Aragon, 1999).

Salinization as outcome of irrigated farming

Despite the important ecological constraints, irrigated agriculture and livestock raising

(cattle, sheep, and goats) have been the most important economic activities in this department

since the mid-19th century. Consequently, the major and probably older human modification of

the natural environment in Jachal has been the implementation of irrigated farming. The areasystematized for irrigation represented 1.2% of the total area of the department, and is

subdivided into five irrigation districts: Jachal (16,437 Has), Niquivil (1,500 Has), Tucunuco

(1,030 Has), Mogna (1,500 Has) and Huaco (1,600 Has). The location of the districts is

displayed in Fig. 1. Irrigated farms represented 89% of the total number of farms in 1988 (1,081),

and all of them used surface sources for irrigation (INDEC, 1990). The irrigation system also

distributes drinking water using a different network. In case of water scarcity and conflicting

needs, domestic consumption has priority over irrigated farming.

Although there are about 22,000 ha with water rights in Jachal, the actual area under

cultivation has been consistently smaller, and present wide interannual variations. For example,

it was 7,161 ha in 1960, 6,440 in 1988, and 2,401 in 1992. Both the small percentage ofsystematized land actually farmed and the annual variations in crop acreage are related to water

management and distribution in three interrelated ways. First, it is not possible to honor all water

rights. The Water Code of San Juan establishes that each farm with water rights is entitled to



8/25

1 litre of water/sec/ha. However, there is not enough water for everyone with water rights in

the districts of Jachal, and an irrigation coefficient is applied. This coefficient indicates what

proportion of the amount fixed by law can practically be served by the real river discharge. For

Jachal, the coefficient is approximately 0.5, which reduces the amount of water to0.5 litre/sec/ha, and it can be even lower in cases of hydrological crisis or emergency such as

drought. Second, the already mentioned naturally high salinity of the Jachal river translates into

farmers requiring 30% more water than normally needed in order to wash the soils. Third, the

use of gravity-based irrigation technology, compounded with inherent topographic steepness,

makes irrigation efficiency at the plot level less than 50% (in other words, more that 50% of the

water is wasted). Efficiency at the level of the distribution network is also low due to the fact that

most of the ditches (212 km out of 300 km) are permeable and water is lost through seepage. 10 In

short, there have a been an underutilization of land resources, compounded with a water

management style that leaves the soils critically exposed to deterioration in case of water

shortages (which are quite common in the area).

Irrigated farming in drylands is commonly associated with salinization (Adamo, 1997;

Mainguet, 1999; Thomas & Middleton, 1993), and Jachal is no exception11. In the irrigated areas

of the department, salinization is the outcome of the natural conditions of water and soils in

combination with inadequate forms of management. One of the most common factors in

increasing the risk and rate of salinization is poor drainage. In 1994 there were 300 km of

irrigation ditches but only 44 km of drainage ditches, rendering a ratio of 0.15: for every 100 m

of irrigation ditches there are only 15 m of drainage ditches. This imbalanced ratio means that it

is easier to transport water to the plot than to take it out.

The magnitude of salinization seems to vary from district to district, although empirical datais scarce.Table 1 displays the distribution of potential agricultural land for the different irrigation

districts in 1971, based on a report elaborated by Harza de Argentina. Out of 40,478 ha of

potential agricultural land, only 51% could be considered productive. There was a wide variation

in the proportion of productive land in each locality, but in every case it was less that 50% of the

total available agricultural land. Overall, salinization affected 24% of the potentially suitable

land, being lower in Jachal and higher in Niquivil and Tucunuco. This report is important

Table 1

Distribution (%) of productive and unproductive land by irrigation district

District Productive land (%) Unproductive land due to

Salinity (%) Slope (%) Urban use (%)

Jachal (31,339 Has) 40 19 41 0.5

Huaco (4,333 Has) 41 35 23 0.5

Mogna (3,218 Has) 35 31 34

Niquivil/Tucunuco (1,588 Has) 16 74 11

Total (40,478 Has) 39 24 37 0.4

Source: After Harza de Argentina (1971).

10 Alfredo Estevez, director of the local Rural Extension Agency of the National Institute of Agriculture Technology,

personal communication, 2001.11 Salinization is a form of chemical degradation related to the accumulation of soluble salts. It is associated with both

natural processes (for example, excessive evaporation in arid climates) and human activities (for example, deficient

irrigation techniques). Among other consequences, salinization reduces biological activity, the content of organic matter

in the soils, and the natural or cultivated vegetation cover (Mainguet, 1999:235).



9/25

because it establishes a suitable starting point to explore environmental change in the irrigated

areas between 1973 and 2001.

Grazing, firewood and the depletion of the natural vegetation cover

The natural vegetation cover has been extensively modified by human activities, especially inand around the settlement areas. It has been used for grazing and firewood, and also replaced by

crops and nonnative species like poplars and eucalyptus planted along the irrigation ditches, a

very distinctive feature of irrigated landscapes in Argentina. It has been also degraded in and

around the settlement, irrigation, and grazing areas (Davire de Musri & Malberti de Lopez

Aragon, 1999; Daz & Luna, 1991; Pannocchia, 1979). As at the end of the 1970s, the

consequences of grazing were apparent in the margins of the irrigation districts, on the

piedmonts, and in the neighbouring mountains (Pannocchia, 1979:86). In 2001, signs of

increasing deterioration of the vegetation cover were visible in the slopes of the mountains

surrounding the irrigation district of Jachal, one of the traditional grazing areas. This decline

could be connected to the reports of progressively more violent run-offs and flooding during

summer storms.

The use of firewood for heating and for cooking in low-income rural and urban sectors is one

of the activities leading to deforestation of native forests in arid and semiarid areas in Argentina

(Morello & Matteuci, 2000). In Jachal, fieldwork indicated that the use of firewood for heating

and boilers in the winter is the norm for all the population. This fact makes firewood gathering a

profitable harvest, especially for owners of fields without water rights. There was some

information pointing to the use of firewood in mining and railroads as other sources of

deforestation. However, the corroboration of this information is difficult due to the lack of

records of past stocks for comparison.

Data and methods

Remotely sensed data were used to explore the characteristics and extension of change in

environmental conditions in Jachals recent past12. Although not without limitations, this

sequence of satellite images allows for observing evolution and change consistently over time.

In developing countries, data about environmental characteristics and dynamics of arid and

semiarid areas are not always readily available. In this sense, the use of remotely sensed data and

techniques provides a partial solution to the lack of information for the study of population-

environment interactions, complementing ground data collection, offering different scales ofspatial coverage, and enabling retrospectives studies (Jensen, 2000; Rindfuss & Stern, 1998)

The study used cross-sectional remotely sensed data for 1973, 1987 and 2001, specifically

three LANDSAT summer (January13) images (MSS and TM). The overlap of the area included

in the processed images and the area of the department of Jachal is displayed in Fig. 4. In order

to have a compatible set of imagery, the preprocessing included re-projection, image-to-image

rectification, and re-sampling.

Processing included a Tasseled Cap Transformation (TCT) (Crist & Cicone, 1984; Jensen,

1996, 2000; Kauth & Thomas, 1976; Tso & Mather, 2001) of the three images to obtain

12 A more detailed account of the preprocessing and processing procedures of the remotely sensed data, including a

discussion about the merits of using vegetation indices as indicators of desertification and the justification of the use of

the Tasseled cap transformation, can be found in Adamo (2003).13 In the Southern Hemisphere, summer lasts from December to March.



10/25

greenness indices for 1973, 1987 and 2001. Unsupervised classification of the resulting images

yielded five final classes for 1973 and 2001. Two classes correspond to vegetated areas: BV,

brigh vegetation, includes the more actively photosynthesizing vegetation, while OV, othervegetation, covers less bright vegetation. Two other classes include soil features: BS, bare soils,

and BBS, bright bare soils, these last likely dryer, with lighter parent materials, or less spotted

with sparse vegetation. The last one, R, is the residual class (mostly rock and shadow)14.

Ancillary dataaerial photographs, topographic maps, photographs and other documentation

were used in the attribution of categories, and in the analysis and interpretation of the results.

In order to detect changes over time, two procedures were performed. The first was image

algebra change detection through image differentiation of the greenness indices (Jensen, 1996).

Differences between images were recoded into five categories, according to the percent increase

Fig. 4. The study area. The figure shows the overlap of the area of the remote sensing images and the department of

Jachal. Although the area of the LANDSAT images does not perfectly match the area of Jachal, it does include the areas

of interest for this exploratory paper, allowing for comparing areas of irrigated agriculture, and of natural and modified

vegetation cover over time. Source: Authors elaboration after Lizana (2001).

14 Accuracy assessments for the unsupervised classifications could not be performed due to the timing of ground data

acquisition and a temporal mismatch between the ancillary sources and the dates of image acquisition. However,

ancillary data were used as the general reference for the attribution of the classes, coupled with the first authors field-

based knowledge of the area.



11/25

or decrease in brightness values: increase (between 100 and 21% higher), some increase

(between 1 and 20% higher), unchanged, some decrease (between 1 and 20% lower) and

decrease (between 21 and 100% lower).

The second procedure was a post-classification comparison change detection used to obtain

from-to class information (Jensen, 1996) between 1973 and 2001, resulting in a change imagemap with 25 classes, one for each pairwise combination of classes. The 25 classes or categories

of the change detection matrix were recoded to nine dominant change categories based upon the

vegetation categories (BV bright vegetation and OV other vegetation). All the aforementioned

procedures were run in ERDAS Imagine for Windows.

Pattern metric analysis was used to asses the configuration of the two image maps for 1973

and 2001, and deduce how the spatial arrangement of the landscape had changed over time to

complement the other landscape characterization methodologies15. The source data (here,

Landsat MSS and Landsat TM) are important for careful interpretation of these metrics, since

the scale of acquisition in part impacts the patterns that are observed and quantified in order to

infer process (Forman, 1995; ONeill et al., 1996; Walsh et al., 2001)16. Metrics for this analysis

were run using the freeware Fragstats (raster data version) (McGarigal & Marks, 1995).

Classified images were exported from ERDAS Imagine as unsigned 16-bit generic binary files,

and run in Fragstats version 3.3 for PC environment17.

Environmental change between 1973 and 2001

Greenness indices

We turn now to the results of the analysis of remotely sensed data to investigate the reportedtrends, including deforestation and crop decline. The histograms of the greenness index derived

from the Tasseled Cap transformation for 1973, 1987 and 2001, displayed in Fig. 5, show a first

indication of change over time. The distribution curves for 1973 and 2001 are very similar, while

the 1987 histogram showed both a lower peak and a displacement toward higher brightness

values. In the greenness index, brightness values are higher where the biomass or total amount of

vegetation is higher or when the growing season (or more accurately, photosynthesis) is at its

peak.

The histograms suggest an overall increase in the amount of vegetation from 1973 to 1987 (in

January 1973, Jachals ecosystem was just starting to recover from a long drought that started

around 1965) followed by an overall decrease between 1987 and 2001. But despite the

similarities, the histograms for 1973 and 2001 are not identical. While the range for 1973 goes

from 55 to 235 with a mean of 94.7, the range for 2001 is displaced toward lower brightness

values, 46199, with a mean of 92.6. This would indicate a small net decrease in vegetation

amount in 2001 relative to 1973.

15 Landscape ecology posits that landscape configuration (spatial arrangement) is as important as landscape

composition (thematic class representation) (Crews-Meyer, 2004). By changing the relative configuration of landscape

elements, so too is the flow of energy and materials on the landscape altered (Forman & Godron, 1986). Landscape

metrics are commonly now calculated from thematic data derived from satellite imagery (Jensen, 2000).16 Scale in this case refers to both grain and extent (Turner, Gardner & ONeill, 2001), and applied to both spatial scale

(extent of study area and resolution of satellite imagery) and temporal scale (range of study period and timing or

seasonality of observation).17 Software and documentation are available online at http://www.umass.edu/landeco/research/fragstats/fragstats.html.

http://www.umass.edu/landeco/research/fragstats/fragstats.htmlhttp://www.umass.edu/landeco/research/fragstats/fragstats.htmlhttp://www.umass.edu/landeco/research/fragstats/fragstats.html


12/25

The image differentiation results presented inTable 2 qualified the type of change observed in

Fig. 5 into five categories. All increases and decreases refer to productive green (non-woody)

biomass as measured by brightness values in the greenness indices. As can be seen, the category

some increase dominated the 19731987 period, while the category some decrease included

more than 90% of the pixels in the 19871901 period. This trend is coincident with what the

histograms in Fig. 5 showed. The numbers for the period 19732001 indicate that 56% ofthe pixels underwent some decrease in the brightness values, with approximately 37% having

0

100000

200000

300000

400000

500000

600000

700000

800000

900000

010

20

30

40

50

60

70

80

90

100

110

120

130

140

150

160

170

180

190

200

210

220

230

240

250

brightness values

counts

1973 1987 2001

Fig. 5. Greenness index histogramsJanuary of 1973, 1987 and 2001. Source: Tasseled cap transformation of the 1973,

1987 and 2001 images.

Table 2

Greenness index image differentiation results (percent of total number of pixels, NZ16,744,000)

Change category 19731987 19872001 19732001

Decreased (20% and more) 0.2 4.5 0.5

Some decrease (119%) 2.4 93.7 56.1

Unchanged 0.5 0.1 6.0

Some increase (119%) 94.3 1.1 37.1Increased (20% and more) 2.7 0.1 0.3

Total 100.0 100.0 100.0

Source: Tasseled cap transformation of the 1973, 1987 and 2001 images.



13/25

experienced some increase. Again, this suggests that there was a small net decrease in biomass in

the study area between 1973 and 2001, and that this decrease has not been linear but fluctuating.

Visual inspection of the maps displaying the greenness index image differentiation results

gives an idea of the spatial distribution of the areas of change. Fig. 6 displays the areas that

changed between 1973 and 2001, following the categories of Table 2. The image is dominatedby areas in the moderate categories (some decrease and some increase), which are located

mostly outside of the irrigation districts, and in segments of the study area with natural

vegetation cover. Areas of some increase cover the mountains on the west and southwest of

Fig. 6. Greenness index image differentiation map, 19732001. Source: Image differentiation of the greenness indexes

from the TCT, 1973 and 2001 images.



14/25

the study area, as well as part of the sierras on the east. Areas of some decrease appear

concentrated in the Travesa on the east, and in the mountains in north of the study area.

Small patches of increase (20% and more) are located in the irrigation districts as well as in

the Travesa on the east. The areas where vegetation decreased 20% or more are mostly

concentrated in the main irrigation districts, especially in Jachal, Tucunuco, and Huaco. There isalso a small concentration in the north mountains, where most of the puestos are located, and on

the east. These results indicate that the sections of the study area where changes between 1973





15/25

and 2001 were more dramatic tend to coincide largely with the areas of settlement, agriculture

and grazing.

Figs. 7 and 8 present a first approximation to the timing of the changes. For the period 1973

1987, displayed in Fig. 7, the areas of some biomass increase dominate the image outside of the

irrigation districts. The areas of 20% or more increase are located in the irrigated districts, aswell as in the natural grazing areas in the mountains on the west and north, and in areas of the

Travesa far east. In the irrigated districts, the areas of increase are intertwined with other

categories. The areas of some decrease are found along the Travesa and in the irrigation





16/25

districts, and to a lesser extent in the north mountains, although these sites are difficult to display

at this scale. Places showing a 20% or higher decrease in biomass are overwhelmingly

concentrated in the irrigated sections. Also, areas of decrease match the location of the puestos in

northern mountains of Jachal.The map for the period 19872001, displayed in Fig. 8, shows a complete dominance of the

areas of some biomass decrease (119%), a reverse trend from the dominance of some increase

in 19731987. Spots of vegetation decrease have multiplied, and some large sections are visible

in the southeast corner of the study area. While these areas and the spots in the western

mountains are new, the proliferation of areas of decrease in the irrigation districts and in the

puestos region appear as a continuation and intensification of the 19731987 trends. The patches

of vegetation increase (20% or more), visible in some of the irrigation districtsJachal, Niquivil

and Huacoand in the eastern Travesa, look very small in comparison.

Classification and change detection

The unsupervised classification of the study area for 1973 and 2001 is shown in Table 3. In

1973, the most frequently observed landscape class was bare soil, which occupied 40.0% or just

over 600,000 ha of the study area. Brighter bare soils covered 11.2% of the landscape. Nearly

one-third of the area was classified as rock/shadow (R), with the remaining 17.4% classified as

vegetation (1.3% bright vegetation and 16.1% other vegetation).

In 2001, the most commonly occurring class was rock and shadow, which covered 40.3% of

the landscape. The second most common class was bare soil, which covered 36.8% of the

landscape or over 555,000 ha. Just over 12% of the landscape was classified as bare bright soil,

with a total of 10.5% of the landscape determined to be vegetation (0.2% or 3721 ha BV and

10.3% or 155,811 ha OV).

Although there are some differences in the distribution of the classes between 1973 and 2001, in

both years bright vegetation represents the smallest proportion, consistent with the small proportion

of Jachal that is best suited for agriculture, the sparse of the natural vegetation cover, and the

abundance of patches of bare soil. There was a decrease in the two vegetationcategories, from 17.3%

of the total area in 1973 to 10.5% in 2001, which agrees with the results in Table 2 and Figs. 68.

Table 4 shows the change detection matrices 19732001 in both percentages of pixels and

area.18

The numbers in the diagonals indicate the percentage of pixels and their corresponding

Table 3

Distribution of the department area by class, 1973 and 2001

Classes 1973 2001

Area (Has) % Area (Has) %

1. Residual (rock and shadow) R 474,251.9 31.4 609,076.9 40.3

2. Bare soils BS 604,659.2 40.0 555,075.4 36.8

3. Bright bare soils (salt, sand, etc.) BBS 168,973.4 11.2 186,195.7 12.3

4. Bright vegetation BV 19,583.3 1.3 3721.4 0.2

5. Other vegetation OV 242,413.1 16.0 155,811.4 10.3

Source: Unsupervised classification of the 1973 and 2001 images.

18 A pixel is a two-dimensional picture element that is the smallest non-divisible element of a digital image (Jensen,

1996:17). In this analysis, pixel dimensions are 30!30 m.



17/25

area that remained in the same class from 1973 to 2001. For example, 77.9% of the pixels

included in R (residual) in 1973 were in that same category in 2001, while the rest (22.1%)appeared distributed in the rest of the classes, indicating the changes between 1973 and 2001.

Looking first at the vegetation categories (BV and OV), over 50% of the area included in BV,

bright vegetation in 1973 was classified as other vegetation in 2001, 13% remained in the

same class, and the rest (32% or 6,200 ha) passed to non-vegetation categories. Regarding OV,

other vegetation, 74% of the area (170,600 ha) changed to the residual category, and 30%

remained in the same. In the case of the BBS, bright bare soils (includes the saline areas), just

under 50% remained in the same category, and 4% (representing 6,800 ha) changed to vegetated

classes. Most of the area in the class BS, bare soils in 1973 remained in the same category, but

3.5% or 22,000 ha changed to OV or BV between 1973 and 2001. Finally, most of the residual

category (78%) remained the same, but about 10% (45,500 ha) changed to other vegetation. In

summary, in 2001 vegetation classes seems to have lost about 102,000 ha or 39% of the area

they covered in 1973, despite the small gain in vegetation from 1973 to 1987 shown by the

greenness indices.

Fig. 9 displays the spatial distribution of the combined 25 from-to categories of the

change detection matrix recoded to nine change classes. This extraction was framed around the

vegetation categories BV (bright vegetation) and OV (other vegetation). Newly vegetated

areas correspond to places that changed from non-vegetation categories (rock, bares soils,

bright bare soils) to vegetation categories (bright or other vegetation). Areas where loss of

vegetation may be presumed are those that changed from vegetation categories to non-vegetation categories.

Two clear sectors are visible in the image: those where changes related to vegetation or

biomass took place and those where they did not. Areas where changes related to vegetation took

Table 4

Change detection matrices: 19732001

Percentages (%)

Classes 2001

1973 1. R 2. BS 3. BBS 4. BV 5. OV

1. Residual (rock and shadow) R 77.9 10.96 1.53 0.01 9.60

2. Bare soils BS 16.05 66.33 13.97 0.09 3.55

3. Bright bare soils (salt, sand, etc.) BBS 3.72 44.49 47.75 0.16 3.87

4. Bright vegetation BV 15.82 12.16 4.08 12.52 55.43

5. Other vegetation OV 54.94 10.09 5.34 0.16 29.46

Area (Hectares)

Classes 2001

1973 1. R 2. BS 3. BBS 4. BV 5. OV Total 1973

1. Residual (rock and shadow) R 369,464.3 51,986.7 7,261.7 26.5 45,512.5 474,251.92. Bare soils BS 97,053.2 401,059.3 84,493.1 574.1 21,479.5 604,659.2

3. Bright bare soils (salt, sand,

etc.) BBS

6,279.1 75,182.4 80,690.9 277.9 6,543.0 168,973.4

4. Bright vegetation BV 3,097.5 2,380.8 798.3 2,452.3 10,854.4 19,583.3

5. Other vegetation OV 133,182.7 24,466.2 12,951.6 390.5 71,422.0 242,413.1

Total 2001 609,076.9 555,075.4 186,195.7 3,731.4 155,811.4 1,509,880.9

Source: Change detection of the unsupervised classification of the 1973 and 2001 images.



18/25

place correspond mostly to the irrigation districts, population centres and grazing areas. Places

where changes were not related to vegetation (in grey in the map) follow the trace of the higher

parts of the mountains and sierras, and most of the Travesa area. Grey areas within the irrigation

districts represent spots without vegetation in 1973 and in 2001.The sections that lost vegetation (in brown and yellow in the map) are concentrated in the

north, west and southwest of the Jachal district, where most of the population lives close or

nearby the grazing zones. There are also visible patches in other irrigated areas, notably between

Fig. 9. Change detection map, 19732001. Source: Change detection matrix, 19732001.



19/25

Jachal and Tucunuco. However, a part of the grazing and puestos areassurrounding Jachal by

the north, on the west of Niquivil-Tucunuco and in the north of the study areaseem to have

remained in the same vegetation class between 1973 and 2001. Other areas where vegetation

categories do not change are scattered in the irrigation districts, notably Niquivil.

The small areas that kept their dense bright vegetation (in red in the map) are hardly visiblein the north and south of the Jachal district, where patches showing relative biomass gains

(OVBV, in light blue in the map) and losses (BVOV, in blue) are also found. Areas that gained

vegetation (in light and dark green) appear concentrated in the irrigation districts (except

Niquivil), and in areas surrounding the puestos on the southwest of the study area. Some small

patches are located in the Travesa, following the water courses.

Landscape metrics and change over time

Table 5 displays the results of the landscape metrics. In 1973, the patch density of the classessuggests that Bright Vegetation (BV) was not only sparsely represented on the 1973 landscape,

but that the size of the patches was very small and dispersed compared with other classes (the

1872 patches of BV were sized and distributed such that the patch porosity was a low 0.12

patches/100 ha). In contrast, the other vegetation (OV) was not only more spatially extensive,

but was distributed in larger patches, with its 25,242 patches averaging 1.67 patches/100 ha. In

addition, BV appeared to be relatively regularly shaped with less edge and simpler geometry

than OV, as the more fragmented OV showed higher edge density akin to bare soil and rock/

shadow. Bright bare soil in particular had an extremely low edge density compared to most other

Table 5

Class metrics, 1973 and 2001

Class CA Pland NP PD ED LSI IJI

Class metrics, 1973

Bright vegetation 19,583 1.3 1872 0.12 1.83 49.31 66.85

Other vegetation 242,413 16.1 25,242 1.67 23.23 178.35 58.61

Bare soil 604,659 40.0 15,017 0.99 27.78 135.55 84.47

Bright bare soil 168,973 11.2 10,012 0.66 11.73 107.93 6.46

Rock and shadow 474,252 31.4 13,034 0.86 21.79 120.08 47.12

Class metrics, 1973

Bright vegetation 3,721 0.2 2037 0.13 0.85 52.32 9.06

Other vegetation 155,811 10.3 76,319 5.05 37.33 357.22 40.53

Bare soil 555,075 36.8 71,495 4.74 55.27 280.63 67.68

Bright bare soil 186,195 12.3 58,743 3.89 28.56 250.14 41.56

Rock and shadow 609,077 40.3 61,488 4.07 64.62 313.20 67.75

Landscape metrics, 1973 and 2001

Year TA NP PD ED LSI IJI

1973 1,509,881 65,177 4.32 43.18 133.65 65.88

2001 1,509,881 270,082 17.89 93.31 287.65 63.94

Class, thematic landscape class; CA, class area, hectares; Pland, percent landscape; NP, number patches; PD, patch

density (patches/100 ha); ED, edge density (m/100 ha); LSI, landscape shape index; IJI, interspersion juxtaposition

index; Year, year of classification; TA, total area; note: IJIZ

fKPm

kZ1eik=Pm

kZ1 eiklneik=Pm

kZ1 eikg=

lnmK

1 whereeik, total length (m) of edge in landscape between patch types (classes) i and k,mZnumber of patch types (classes) present

in the landscape, including the landscape border, if present. LSIZ eimin eiwhere ei, total length of edge (or perimeter) of

class i in terms of cell surfaces, includes all landscape boundary and background edge segments involving class i. min ei,

minimum total length of edge (or perimeter) of class i in terms of number of cell surfaces.



20/25


21/25

(e.g. converting BV to OV, due to water shortages, purposive land conversion, or natural

succession to older, less productive vegetation). The latter is strongly supported by the change

analysis.

Considering the above findings in light of the landscape metrics provides additional insight

into the changing composition and configuration of these landscape elements. In 1973, therewere just over 65,000 patches on the landscape; by 2001, this number had increased more than

fourfold to over 270,000 patches. Obviously the landscape became patchier, with patch density

or porosity also increasing fourfold. Edge density doubled between 1973 and 2001, as did LSI (a

measure of shape complexity). Interestingly, however, IJI remained relatively constant over the

period, suggesting that while the landscape was becoming more fragmented, the classes

themselves were on average not becoming more interspersed, meaning that the fragmentation

resulted in a new landscape configuration in terms of shape but not, on average, in terms of

thematic composition (though some individual classes did show changes in IJI, these were less

spatially prevalent and thus did not impact the landscape metrics as much).

Discussion

The results from the analyses of the remote sensing data indicate several changes between

1973 and 2001. While it is certainly plausible to suggest that these changes might be artifacts

of the classifications or differing data sources, the magnitude of the changes and the

consistency among the different landscapes metrics evaluated here instead point to a

landscape that is undergoing a reorganization that, as with many arid landscapes, is substantial

but subtle.The results show cycles of vegetation decline and regeneration that are likely associated with

the variations in moisture availability in the department. Differences between 1973 and 1987 are

probably linked, at least partly, to the recovery from the drought that struck the department in the

second half of the 1960s. These fluctuations were found mostly outside the settlement areas

(irrigated districts and grazing zones). There is also evidence of a net decline in biomass between

1973 and 2001, despite the recuperation in the period 19872001, and a slight increase of areas

of bright and bare soils (the BBS and BS classes). This may be a first indication of land

degradation leading to desertification, since in this area bright bare soils are typically indicative

of salty land unsuitable for agriculture.

The pattern of the changes appears to be different in the irrigation districts, grazing areas andsurrounding zones, as shown by the greenness and change detection maps. The several areas

where change in the greenness index was classified as 20% or more were consistently located in

and around irrigated places. Also, the irrigation districts and to a lesser extent the grazing areas

displayed a very fragmented or patched pattern in the distribution of the type of changes. These

may be indicating not only more changes but also increasingly aggressive ones.

Overall, the review of the environmental framework and the forms of management taken

with the results of the analysis of remotely sensed data indicate that both aridity and

desertification are behind environmental change in Jachal. The relative influence of human

and natural factors in land degradation in the department are intertwined, and the

environmental constraints have interacted with the forms of use and management of naturalresources for a long time, reinforcing each other. For example, while the irrigation system

buffers some of the effects of aridity, the particular irrigation system at work (gravity)

contributes to land degradation by increasing salinization. In turn, the typical limitations of



22/25

gravity irrigation systems are exacerbated by the salty quality of the waters of the Ja chal

river.

A basic interpretation is that the modifications in the irrigated and populated areas are more

likely due to social processes, while changes in the rest of the department are a response to

natural processes. Increases in salty areas could have been promoted by a particularly drysummer if located in the eastern Travesa, but if this change were found in the irrigated districts

it would very likely be due to inadequate irrigation practices.

Changes in land use are likely behind some of the changes detected between 1973 and

2001. Tax incentives for crop diversification, implemented during the 1990s, substantially

increased the amount of land in olive orchards in the Jachal and Huaco irrigation districts.

Olive trees require between 5 and 8 years to grow, and in between the soil looks bare

because of the very low coverage of the young trees. Bare soils are more exposed to wind

and water erosion.

Another relevant social process is population redistribution. The 1991 and 2001 Argentinas

censuses of population indicated a sustained decline of population in the south west section of

the study area. This decrease could be an indication of the abandonment of puestos in this

grazing sector, one of the areas where an increase in vegetation was detected. Conversely, the

signs of vegetation decline around the district of Jachal (notably on the north and east) may be

the result of overgrazing, firewood gathering, and a relatively recent expansion of the cultivated

area. Population settlement has increased in this section of the district, which is relatively more

sensitive to land degradation because of the slopes. Finally, the small patches of vegetation

decrease in the northern mountains coincide with the location of Gualcamayo, one of the oldest

mountain settlements of Jachal.

It is important to remember that the images, each representing a fixed point in time, may beregistering just a stage of an ongoing process. It was observed during fieldwork in Mogna that

vegetation cover in areas under waterlogging (raising of the water table) and salinization went

through different phases before being killed by the salt. In one of those phases, vegetation

appears dense and green, but the only species left at this point are halophytes (jumes and zampas

are their local name) not suitable to be used as forage. This aspect is relevant for the

interpretation of remotely sensed data, since green vegetation may sometimes be indeed an

indication of land degradation. In these cases, it is necessary to pay greater attention to location,

patterns and shapes, and ground verification ancillary data.

Conclusions

This exploratory analysis suggests a complex panorama with respect to environmental

hazards in Jachal, as well as their evolution and nature. Overall, aridity is simultaneously the

typical characteristic, the primary environmental constraint and the main hazard of

the department of Jachal. Aridity means that water shortages in general as well as seasonal

and inter-annual variations in water availability are the norm. Climatic events such as

droughts, floods, and hailstorms add to the hazard posed by the characteristics of the arid

climate. The drainage is limited, with multitude of occasional and temporary courses and just

two permanent rivers, the Jachal and the Huaco, which have structured population settlement.

Soil quality is not optimal and natural vegetation cover is low, although extremely modifiedin the irrigated areas.

However, aridity is not the only process at work in this landscape, and its effects are

complicated by both natural resources management issues and other social issues related



23/25

to agriculture and settlement in arid lands. In Jachal, the main natural resources are water,

land, and to a lesser extent the natural vegetation cover, and irrigated farming and

ranching have profoundly modified the natural landscape. Seemingly, the local and

historical forms of management of this arid environment have led to a number of

environmental problems, among them salinization, waterlogging, deforestation, and soilinfestation. But the incidence and significance of the problems presents marked spatial

and temporal variations.

The physical and biological environment may be regarded as part of the context that shape

social phenomena (Rindfuss & Stern, 1998:3). In this sense, environmental change and

degradation present a critical issue in arid and semiarid agricultural areas because they affect, in

different ways, the economic basis of the livelihood strategies of the population by changing the

circumstances under which decisions are made. The long-term population-environment

relationship in Jachal suggests that the population has adapted to the type of hazards represented

by the natural constraints and the environmental problems associated with the management of

natural resources. The question remains if they can continue to adapt should these processes

continue a linear rather than cyclical evolution.

Although the evidence presented in this paper does not indicate a generalized process of

desertification in the study area, it does allow for the conclusion that land degradation should be

included among the environmental hazards in Jachal. Even if a rampant desertification process is

not yet apparent, it is likely that vegetation loss in the irrigation districts, grazing areas, and areas

of firewood gathering are pointing to incipient and currently localized processes of land

degradation that could eventually trigger desertification if they overcome the resilience of this

agro-ecological system.

Acknowledgements

While preparing this paper, S. Adamo was supported by a fellowship from the Andrew

W. Mellon Foundation Urbanization and Migration Center Grant at the Carolina Population

Center. The original research was supported by The Population Council, the Andrew W. Mellon

Foundation, and the Center for the Study of Urbanization and Internal Migration in Developing

Countries of the Population Research Center, University of Texas at Austin.

References

Abraham, E. (2003). Introduccion. In E. Abraham, D. Tomasini, & P. Maccagno (Eds.), Desertificacion: indicadores y

puntos de referencia en America Latina y el Caribe (pp. 1115). Editores: Mendoza Zeta.

Adamo, S. (1997). La relacion poblacion-ambiente en el Distrito de Riego Valle de El Carrizo (Sinaloa), a partir de un

enfoque de mediaciones. In C. Rabell (Ed.), Los retos de la poblacion (pp. 1956). Mexico, DF: FLACSO/Juan

Pablos Editor.

Adamo, S. (2003). Vulnerable people in fragile lands: migration and desertification in the drylands of Argentina. The

case of the Department of Jachal. PhD Dissertation. University of Texas, Austin.

Agnew, C. (2002). Drought, desertification and desiccation: The need for further analysis. Geography, 87(3), 256267.

Allub, L., & Guzman, L. (2000). Las estrategias de sobrevivencia de los pequenos productores rurales de Jachal, SanJuan, Argentina. Estudios Sociologicos, 18(52), 125165.

Blaikie, P., & Brookfield, H. (1987). Land degradation and society. London: Longman.

Cabrera, A. (1958). Fitogeografa. In F. Aparicio, & H. Difrieri (Eds.), La Argentina. Suma de geografa (Vol. III) (pp.

101207). Buenos Aires: Peuser.



24/25

Centro Regional del Agua Subterrranea (1972). Investigacion del agua subterranea en los Valles de Jachal-Niquivil y

Huaco, provincia de San Juan. San Juan. Unpublished manuscript.

CEPAL/GTZ. (2003). Indicadores de impacto socioeconomico de la desertificacion y degradacion de tierras. Paper

presented at the Taller Nacional Evaluacion de Indicadores de Impactos Socioeconomicos de Desertificacion y

Degradacion de Tierras. Buenos Aires, November 19th.

Clarke, J., & Noin, D. (Eds.). (1998). Population and environment in arid regions (pp. 118). Paris: UNESCO/Partenon

Publishing Group.

Cornejo, R. (1997). Clima de Jachal. Unpublished manuscript.

Crews-Meyer, K. (2004). Agricultural landscape change and stability: Historical patch-level analysis. Agriculture,

Ecosystems, and Environment, 101, 155169.

Crist, E., & Cicone, R. (1984). A physically-based transformation of thematic mapper data the TM tasseled cap. IEEE

Transactions on Geoscience and Remote Sensing, GE-22(3), 256263.

Davire de Musri, D. & Malberti de Lopez Aragon, S. (1999). Historia de Jachal. San Juan: Gobierno de San Juan/H.

Legislatura/Instituto de Historia Regional y Argentina H. D. Arias.

Deng, L. (2000). Integrating drylands issues into country assistance strategies: a review. Washington: The World Bank.

Daz, M. & Luna, S. (1991). Un analisis del proceso de acumulacion en el Departamento de Jachal. B.A. Thesis.

San Juan, Universidad Nacional de San Juan.Dregne, H. (2002). Land degradation in the drylands. Arid Land Research and Management, 16(2), 99132.

Dregne, H., & Chou, N. (1992). Global desertification dimensions and costs. In: Texas Technological University.

Degradation and restoration of arid lands. Lubbock (TX).

Forman, R. (1995). Land mosaics: The ecology of landscapes and regions. Cambridge: Cambridge University.

Forman, R., & Godron, M. (1986). Landscape ecology. New York: Wiley.

Harza de Argentina (1971). Estudio de prefactibilidad tecnico-economico de una presa de embalse sobre el R o Jachal.

Provincia de San Juan, Buenos Aires: CFI.

Hewitt, K. (1997). Regions of risk A geographical introduction to disasters. Essex: Addison Wesley/Longman.

INDEC (1990). Censo nacional Agropecuario 1988. Resultados generales. Provincia de San Juan. Buenos Aires.

INTA (1990). Atlas de suelos de la Republica Argentina. Escala 1:500,000. Buenos Aires: SAGyP./INTA.

INTA/Aeroterra (1995). Atlas de suelos de la Republica Argentina. Buenos Aires: ArgenINTA.

International Organization for Migration (1992). Migration and the environment. Background Paper. Nyon.Jensen, J. (1996). Introductory digital image processing A remote sensing perspective. Upper Saddle River: Prentice-

Hall.

Jensen, J. (2000). Remote sensing of the environment An Earth resource perspective. Upper Saddle River: Prentice-Hall.

Kauth, R., & Thomas, G. (1976). The tasseled cap a graphic description of the spectral-temporal development of

agricultural crops as seen by LANDSAT. Paper presented at the Symposium on Machine Processing of Remote

Sensed Data. Purdue University: West Lafayette.

Kiesling, R., dir. (1994). Flora de San Juan. Republica Argentina. Buenos Aires: Vazquez Mazzini Editores.

Knerr, B. (1998). Impacts of labour migration on the sustainability of agricultural development in arid regions. In J.

Clarke, & N. Noin (Eds.), Population and environment in arid regions (pp. 211246). Paris: UNESCO.

Little, P. (1994). The social context of land degradation (desertification) in arid lands. Population and environment:

rethinking the debate (pp. 209251). Boulder (CO): Westview Press.

Lizana, C., dir. (2001). Atlas socioeconomico multimedia San Juan; 2002. San Juan CEFOCCA. Preliminary version.

Mainguet, M. (1999). Aridity: droughts and human development. Berlin: Springer.

McGarigal, K. & Marks, B. (1995). FRAGSTATS: spatial pattern analysis program for quantifying landscape structure.

USDA For. Serv. Gen. Tech. Rep. PNW-351.

Middleton, N., & Thomas, D. (1997). World atlas of desertification. London: UNEP/Arnold.

Morello, J. (1958). La provincia fitogeografica del monte. Opera Lilloana, 2, 1155.

Morello, J., & Matteucci, S. (2000). Singularidades territoriales y problemas ambientales de un pas asimetrico y

terminal. Realidad Economica 169, 7096.

Murray, S., Burke, L., Tunstall, D., & Gilruth, P. (1999). Drylands population assessment II. UNSO.www.undp.org/seed/

unso/concepts&programs/pub-htm/dryland-population.pdf .

Noin, D., & Clarke, H. (1998). Population and environment in arid regions of the world. In J. Clarke, & D. Noin (Eds.),

Population and environment in arid regions (pp. 118). Paris: UNESCO/Partenon Publishing Group.ONeill, R., Hunsaker, C. T., Timmins, S., Jackson, B., Jones, K., Ritters, K., et al. (1996). Scale problems in reporting

landscape pattern at the regional scale. Landscape Ecology, 11, 169180.

Pannocchia, S. (1979). Geograf a agraria de la zona bajo riego del departamento de Jachal (San Juan). Cordoba:

CEIFAR.

http://www.undp.org/seed/unso/concepts&programs/pub-htm/dryland-population.pdfhttp://www.undp.org/seed/unso/concepts&programs/pub-htm/dryland-population.pdfhttp://www.undp.org/seed/unso/concepts&programs/pub-htm/dryland-population.pdfhttp://www.undp.org/seed/unso/concepts&programs/pub-htm/dryland-population.pdf


25/25

PROSA. (1988). El deterioro del ambiente en la Argentina (suelo, agua, vegetacion, fauna). Buenos Aires: FEBIO.

Rindfuss, R., & Stern, P. (1998). Linking remote sensing and social sciences: the need and the challenges. In D.

Liverman, People and pixels. Linking remote sensing and social science (pp. 127). Washington: National Academy

Press.

SAGyP (Secretara de Agricultura, Ganadera y Pesca) & CFA (Consejo Federal Agropecuario). (1995). El deterioro de

las tierras en la Republica Argentina. Buenos Aires: SAGyP.Strahler, A., & Strahler, A. (1992). Modern physical geography. New York: John.

Thomas, D., & Middleton, N. (1993). Salinization: new perspectives on a major desertification issue. Journal of Arid

Environments, 24, 95105.

Thomas, D., & Middleton, N. (1994). Desertification: Exploding the myth. Chichester: Wiley.

Tso, B., & Mather, P. (2001). Classification methods for remotely sensed data. London: Taylor & Francis.

Turner, M., Gardner, R., & ONeill, R. (2001). Landscape ecology in theory and practice: pattern and process. New

York: Springer.

UNCCD (2004). FAQ. http://www.unccd.int/knowledge/faq.php#answer2.

UNEP (1992). World Atlas of Desertification. London: Arnold.

United Nations (1994). Elaboration of an international convention to combat desertification in countries experiencing

serious drought and/or desertification, particularly in Africa. Final text of the Convention. www.unccd.int/conven-

tion/text/pdf/conv-eng.pdf.

UNSO (United Nations Office to Combat Desertification and Drought) (1997). Aridity zones and dryland populations: an

assessment of population levels in worlds drylands, New York.

Walsh, S., Crews-Meyer, K., Crawford, T., Welsh, W., Entwisle, B., & Rindfuss, R. (2001). Patterns of change in land

use/land cover and plant biomass: separating intra- and inter-annual signals in monsoon-driven Northeast Thailand.

Remote Sensing and GIS Applications. In A. Millington, S. Walsh, & P. Osborne (Eds). GIS and remote sensing

applications in biogeography and ecology. Boston: Kluwer Academic Publishers.


Documents

Aridity Argentina