SCOPE 53 - Methods to Assess the Effects of Chemicals On Ecosystems

10.1 INTRODUCTION

Arid ecosystems possess a number of characteristics that form the basis for their response to natural and anthropogenic stress. Of these characteristics, none is more dominant than precipitation and water availability (MacMahon, 1981; Smith and Morton, 1990). Low annual precipitation (generally less than 25 cm/year) results in plant communities typically dominated by shrubs and grasses. Furthermore, spatial and temporal variability in precipitation and soil moisture result in heterogeneity in plant and animal community structures. Desert soils are generally infertile and nutrients tend to accumulate under shrubs rather than between them (Smith and Morton, 1990; McAuliffe, 1988).

Arid ecosystems are remarkably stable when compared to other ecosystems. Succession often appears non-existent or is difficult to document. Although little species turnover may exist over large spatial scales, considerable changes can be observed in species dominance within a local area. Furthermore, the rate and degree of natural and anthropogenic disturbance can have an effect on species turnover and succession in arid ecosystems (McAuliffe, 1988; Turner, 1990; Webb et al., 1987).

Overexploitation of arid land resources, combined with low and erratic precipitation and relatively infertile soils, have led to severe land degradation and reduced productivity. Loss or decline in potential or actual biological productivity (desertification) is one of the most significant global environmental issues facing mankind (Speth, 1988). Because of low and erratic precipitation and relatively infertile soils, arid ecosystems recover very slowly following disturbance (Webb et al., 1987).

Several human activities have affected, and continue to affect, the condition of arid ecosystems. These include grazing, land conversion (e.g., to agriculture), water development, salinization, recreation (off-road vehicle use), urbanization, and mining. Grazing by domestic livestock has had a dramatic effect on plant species composition; overgrazing often reduces species richness and increases aridity (e.g., decreases soil moisture), soil erosion, and invasion of exotic plant species (UCAR, 1990). Land conversion typically reduces the amount of native habitats and often results in development of limited water supplies. Agricultural conversions lead to the development of a monoculture, increases in soil bulk density, and decreases in infiltration. These changes may lead to increased soil salinity that can eventually preclude plant growth. Additionally, increased salinity combined with mobilized toxic trace elements from agricultural pesticides and fertilizers may have a dramatic effect on plant species composition (Kepner and Fox, 1991).

Rapid urbanization and development of arid regions has raised a concern about the declining air quality and its associated effects on arid environments. Of greatest concern are nitrogen oxide and sulphur dioxide emissions from automobiles and industry, and ozone. Unfortunately, existing air monitoring networks provide poor coverage of air pollutants within arid lands; therefore, concentrations and deposition rates of airborne pollutants is largely unknown. Additionally, few studies have addressed the impacts of airborne chemicals on arid ecosystems. Thompson et al. (1984) showed that certain desert annuals were extremely sensitive to ozone at concentrations below current standards. Hill et al. (1974) found that many native desert plants were highly resistant to injury from SO₂ and NO₂, although at least one plant (Indian ricegrass) showed injury at lower concentrations of SO₂ (1 ppm vs. 4 ppm) when soil moisture was increased. This observation suggests that a concentration that is safe during relatively dry years may cause injury in years with high rainfall. Dawson and Nash (1980) studied the impacts of a smelter on a desert shrub community in Arizona, and found that large shrubs were unaffected by smelter emissions but that shallow rooted plants were. They attributed lack of impact on large shrubs to their reduced number of stomata and the short length of time stomata are open (as compared to plants in more mesic environments). Impact on shallow-rooted species was attributed to chemical changes in the upper layer of soils.

Arid zones are spread over wide ranges of the world, between the latitudes 10-50°N and 10-40°S. Arid zones are particularly prominent at latitude 30°N and 30°S on account of global weather patterns. A "convection cell" in the atmosphere results from the excessive heating at the equator which causes hot air to rise. The rotation of the earth causes a downward convection of the air at around 30°N and 30°S. The rising air at the equator expands and cools, thus losing its water vapour content and forming the typical cloud cover of the equatorial regions. When the air is convected downwards near latitudes 30°N and 30°S, it becomes drier and land areas in this latitude comprise the primary desert belt of the planet.

In addition to this anticyclonic activity, arid zones may be caused by continentality (often either exacerbating or exacerbated by anticyclonic activity), topography, and cold ocean current. The one factor that they have in common is a lack of moisture in normal climate conditions (UNEP, 1992). Taken together, these areas comprise approximately 37 percent of the land area of the earth.

Numerous methods exist to determine the effects of natural and anthropogenic chemicals in arid ecosystems. This chapter is mot meant as an exhaustive treatise encompassing all chemical stressors and all methods, but rather to present an overview of the basic considerations involved both from the standpoint of the stressors and from the perspective of the techniques. Specifically, the methods will be briefly outlined for the assessment of the effects of chemicals that are largely unique to arid environments on soils, and will discuss techniques to assess land degradation as a function of climate and anthropogenic stressors.

10.2 METHODS TO ASSESS THE EFFECTS OF CHEMICALS ON SOILS

Of the desert and semi-desert areas, approximately 15 x 10^-6 km² is characterized by a lack of leaching resulting in an accumulation of soluble salts, gypsum, carbonates, and sodification in the soils (UNFAO, 1971-1981).

The major chemical stress on arid-zone ecosystems is increased salinization in soil and water which may also lead to sodification of soils. The diminished leaching of arid zone soils combined with increased inputs of salts through urbanization and irrigation, and the limited removal of salts, results in a slowly (tens to hundreds of years) increasing salinity of soil and local water resources.

Increased salinity adversely affects plants through increased osmotic pressure and decreased water availability. Increased concentration of exchangeable sodium in soils, caused by the increased relative concentration of Na⁺ in saline soil solutions, is causing soil-structure destabilization, decreased hydraulic conductivity, crust formation, and increased susceptibility to water erosion.

A well-established system of methods for salinity assessment in soils and water has been developed by the USDA Salinity Laboratory in Riverside, California. Originally, it consisted of analysing the concentration of the major soluble ions (Ca, Mg, Na, K, Cl, SO₄, HCO₃/CO₃) in soil extracts and in water. In recent years, with the advent of modern, rapid, multielemental analytical techniques such as AA and ICP, and the increased demand for the assessment of a wider range of environmental contaminants, more comprehensive elemental analysis schemes are conducted including nitrate, essential trace elements (Fe, Mn, Zn, Cu, B), and potentially toxic trace elements (e.g., Cd, Cr, Ni, As, Se, Hg).

The availability of comprehensive elemental concentration data in soil solutions and waters supplies the needed data for chemical speciation modelling in solution and prediction of precipitation-dissolution, sorption-desorption and partitioning-transport processes of elements in the soil system. Advanced computer codes using thermodynamic equilibria have been employed to fully characterize the chemistry of soils and water (Sposito and Mattigod, 1979). These are still in the developmental stage.

10.2.1 ANALYTICAL PROCEDURES FOR SOIL SOLUTIONS AND WATERS

10.2.1.1 Soil solution extraction

Since the extraction of a soil solution at the field-moisture range is a difficult task, the customary procedure is to conduct the extraction at a higher (water to soil) ratio which is determined by the soil itself to ensure appropriate representation. The saturated soil-paste (SP) is the preferred condition (Rhoades, 1982). In this method, the soil is brought to saturation with distilled water, the paste is permitted to reach chemical equilibrium and the soil solution is extracted by applying a vacuum.

10.2.1.2 Total salinity

A simple method to assess total salinity in solution has been the gravimetric method of total dissolved solids (TDSs). More recently, the measurement of electrical conductivity (EC) at standardized conditions has replaced the gravimetric method. The EC method is widely accepted for routine work due to its rapidity and good correlation with total concentration of charged ions in soil extracts and water.

10.2.1.3 Electrical conductivity of the solution

The electrical conductivity of the extract is used to estimate the following general solution concentration parameters:

1. total cation (or anion) concentration, meq/L = 10 x EC (ds/m, 25°C);
2. total salt concentration, mg/L = 640 x EC (ds/m, 25°C);
3. osmotic pressure, bars at 25°C = 0.39 x EC (ds/m, 25°C).

10.2.1.4 Analysis of ionic constituents

Several methods are available for the analysis of the ionic constituents in soil extracts. Where available, the use of ICP-AES or ICP-MS enables the rapid, simultaneous analysis of many of the major, minor, and trace components in the soil solution. Single-column ion chromatography has also been tested and optimized for the sequential simultaneous analysis of the major and minor ionic constituents in soil extracts (Nieto and Frankenberger, 1985a, 1985b).

The following compilation represents those methods routinely used in advanced field laboratories. Details of the technical steps can be obtained from the respective instrument manuals.

10.2.1.5 Soluble major cations

Atomic absorption is customarily used to analyse Ca²⁺ (0-0.4 meq/L concentration range). Mg²⁺ (0-0.1 meq/L) and K²⁺ (0-0.1 meq/L) (Rhoades, 1982). Soluble major anions include:

1. Chloride: automatic chloridometer.
2. Carbonate/bicarbonate: automatic potentiometric titrator (Rhoades, 1982).
3. Sulphate: sulphate analysis has not yet reached the level of instrumental sophistication as routinely available for most other soluble constituents in arid waters and soils. Available methods include turbidimetric analyses (Rhoades, 1982), and ion chromatography (e.g., Nieto and Frankenberger, 1985a).
4. Nitrate: a selective-ion electrode is used to measure nitrate activity potentiometrically using an expanded scale potentiometer (Keeney and Nelson, 1982).
5. Boron: spectrometric analysis at 420 nm.

10.2.1.6 Analytical procedures to assess solid phase composition

Soils of arid lands are modified by anthropogenic inputs of chemicals in much the same way as in other ecosystems. A process, somewhat unique to arid regions, is that of sodification or the increase in the proportion of exchangeable Na+ above about 15 percent of the cation exchange capacity (CEC) of the soil. Pollution by heavy metals (e.g., Cd, Pb, Cr, Ni) is observed due to atmospheric fallout, municipal waste disposal, use of fertilizers ( e.g., Cd in P-containing fertilizers), and prolonged use of reclaimed sewage effluents. Due to a general lack of leaching, low organic matter content, and high pH of arid soils, the retention of potentially toxic trace elements in the top layer of the soil is enhanced. This situation increases the probability of heavy metal introduction into the food chain or their direct ingestion by animals or humans.

10.2.1.7 Exchangeable cation composition of soils

Exchangeable ions are electrostatically bound to charged sites in the soil mineral and organic components. They are readily exchanged and displaced by excess of any neutral salt. In arid zone soils, the major contributor to the soil CEC are the mineral components, mostly clay minerals such as montmorillonite, illite, palygorskite, and kaolinite. The exchangeable ions are Ca²⁺, Mg²⁺, Na²⁺, and K⁺, usually appearing in that order of abundance. Increased proportion of Na⁺ causes destabilization of the soil structure and adverse effects on its permeability to water.

The composition of exchangeable ions is determined after their displacement by a relatively concentrated salt solution. The displacing solution of choice is lN ammonium acetate (NH₄C₂H₃O₂) (Thomas, 1982), since its cation is not part of the exchangeable ions in nature and it presents only limited interferences in the determination of the four exchangeable ions by flame photometry and/or atomic absorption spectrometry. Some dissolution of CaCO₃ or CaSO₄·2H₂O when present in the soil by the displacing solution, may cause erroneous estimation of exchangeable Ca²⁺, and the method should be avoided in such soils, or calcium content be determined by the difference between the total cation exchange capacity and the sum of exchangeable (Mg + Na + K).

10.2.1.8 Extractable ("available") trace elements by DTPA

The DTPA (diethylenetriaminepentaacetic acid) method was developed to measure the available concentration of essential trace elements-Fe, Mn, Zn, and Cu. It may also be appropriate for the estimation of available Ni and Cd in polluted soils (Baker and Amacher, 1982). This method attempts to extract the labile pool of the metals in the soil which is readily available to plants growing in it. Further calibration of this method is required, since it is not an equilibrium extraction procedure and highly dependent on the conditions of extraction.

10.3 REMOTE SENSING TECHNIQUES

Arid ecosystems provide an interesting set of problems and issues with which to study the effects of environmental chemistry. A unique geochemistry exists over much of arid regions related primarily to the low rainfall and high evaporation. With clear skies and low cloud cover, remote sensing techniques might be used to identify, assess, and monitor the effects of chemical exposure in arid environments. Finally, with moisture in short supply, many arid ecosystems are near the edges of their ecological tolerances; as such, subtle changes in the environmental chemistry may bring about highly significant changes in attributes of the ecosystems.

Remote sensing systems can be used to identify, assess, characterize, map, and monitor contaminant exposure. This section is not meant to serve as a complete guide to all remote sensing systems or capabilities, but rather those that are felt to have direct use or potential use to analyze environmental stress in arid and semi- arid ecosystems. Remote sensing technologies that can be used for environmental assessment may be readily available ("off-the-shelf") or may be developmental. Clearly, the experience and knowledge of the person doing the assessment is a major factor in determining the use of a particular system.

10.4 BASIC REMOTE SENSING CONCEPTS AND PRINCIPLES

The use of remote sensing for the characterization and assessment of environmental stress is predicated on some basic assumptions and principles. The spectral properties of materials can be used not only for identification as to type, property, or composition, but also for characterization, that is an assessment of some sort of condition or quantification of properties of the material. Spectral remote sensing can be used to derive information about surface materials based on their reflectance and/or emittance behaviour. The technology can also be used to discriminate subsurface (metres to 10s of metres) phenomena provided the phenomena have affected (through capillary action, for example) the near surface environment.

The spectral reflectance and spectral emittance characteristics of vegetation, soils (and rocks), and water in different wavelength regions are a result of the chemical and physical properties of these materials. Thus, a strategy employing remote sensing for discriminating surface or near-surface phenomena must take advantage of the unique spectral manifestations of the materials, within specified wavelength regions, within certain temporal guidelines, analyzed by a specific set of techniques, and perhaps integrated with other methods or techniques.

The use of sensors to discriminate and characterize surface and subsurface materials is not restricted solely to systems operating within the electromagnetic spectrum. Sensors (e.g., radiation sniffers) that may be considered to be "direct sensors" can be considered in this context. Other considerations for a remote sensing-based strategy of the assessment of environmental chemistry in arid ecosystems include GIS (geographic information systems) integration, techniques for data analysis, and vegetation response.

10.4.1 SENSORS AND CAPABILITIES

10.4.1.1 Aerial photography

This process is used to characterize the site setting (including vegetation, drainage pattern, soils, etc.) for change detection, and for direct visual examination of contaminants (directly as in oil and chemical spills, and indirectly as in discarded barrels and soil and vegetation response). Aerial photography has the advantage of having very high spatial resolution, but limited spectral resolution.

10.4.1.2 Multispectral imagery (less than 2.4µm)

Examples include the lower spectral resolution Landsat satellite sensors (multispectral scanner, MSS, and thematic mapper, TM), the French SPOT satellite, and the Daedalus (Landsat TM simulator) airborne scanner, and the higher spectral resolution (24 channel) (hyperspectral) Geoscan airborne scanner. These available systems are useful for site characterization (the satellite systems are primarily useful for larger area site characterization) and for specific (but fairly coarse scale) soils and vegetation responses to the geochemistry. While the Landsat sensors have fairly coarse resolution (30 m and 80 m), their high temporal resolution characteristics and digital data format allows them to be quite useful for change detection studies. Thematic mapper (TM) simulators typically have the same band passes as are on the TM but are housed on aircraft. One advantage of these systems is that they allow for imaging convenient to a mission need, and that spatial resolution can be improved (sometimes considerably) over the coarser scale TM. The French SPOT (systems probatoire de I'observation de la Terre) satellite provides a critical link between the poorer spatial resolution but more appropriate spectral configuration of the Landsat MSS and TM and the very high spatial resolution of aerial photography.

10.4.1.3 Hyperspectral sensors

Hyperspectral remote sensing is predicated on the need for narrower band sensors to discriminate and characterize materials with greater precision than the coarser Landsat bandpasses. The Geoscan sensor is a 46 channel sensor (with 24 channels available for a given mission) operating in the visible, near-infrared, and thermal infrared portions of the spectrum. Bandpasses, while narrower than those of Landsat or SPOT, are nevertheless fairly coarse (20 to 40nm). Recent research describes the need for 10-20 nm for characterization of minerals in the 2.0 to 2.4 µm spectral region and for 10 nm or less for discriminating vegetation spectral responses to geochemical conditions. The CASI (compact airborne spectrographic imager) is a commercially available system with very fine spectral resolution (2 nm) but unfortunately operates only in the 0.4-1.0 µm region.

10.4.1.4 Thermal infrared sensors

Thermal sensors directly measure the emitted thermal energy from objects (including the earth's surface). A variety of these sensors exist and range from broad-band temperature measuring devices to narrower-band multispectral instrumentation (e.g., the thermal capabilities of Geoscan).

10.4.1.5 Microwave systems

Passive systems can measure surface and subsurface temperature and soil moisture. Active microwave (RADAR) can differentiate surface and subsurface disturbances.

10.4.1.6 Lasers

The principle of laser remote sensing involves the projection of a narrow beam of coherent visible or near-infrared light and then measurement of the reflected radiation. Laser-induced fluorescence involves the measurement of emission spectra at specific wavelengths. The measurement of emission spectra near 690 nm and 740 nm can be diagnostic for monitoring plant stress.

10.4.1.7 Vegetation spectral response

Numerous researchers have made use of the response of vegetation to geochemical conditions to analyze the near-surface environment. Vegetation is known to respond to anomalous environmental geochemistry in three principal ways: taxonomically, structurally, and spectrally. The taxonomic response involves species of community differences in areas having anomalous environmental geochemistry. The structural response involves growth pattern (or "phenologic") and physiognomic indicators. These might include stunting, more open vegetation, or affected flowers and fruits. The spectral response may involve the spectral manner in which the previous two are manifest. Researchers have described a movement of the red edge of plant reflectance to shorter or longer wavelengths as being indicative of stress. One critical issue involves the separation of moisture stress from chemical stress. Research conducted at the Desert Research Institute, the USGS, the University of California at Davis, and elsewhere suggests that this separation is feasible.

Several other systems have been developed, which, although not currently operational, have considerable potential to assess environmental contamination. Chief among these are airborne and satellite hosted hyperspectral imaging Spectrometers. These systems (the AVIRIS is a current prototype) allow for very fine spectral resolution (2-10nm) over the 0.4-2.4 µm range and allow for highly diagnostic information on surface materials.

Ground techniques are frequently used to examine vegetation affected by environmental chemistry. The use of these techniques includes various types of spectroradiometers that may either duplicate the bandpasses of the airborne and satellite sensors or, in some cases, provide considerably greater spectral detail. These spectra were acquired by a Personal Spectrometer-2 (PS-2), an instrument operating in the visible and near-infrared which can acquire spectra having a resolution of 2 nm. The top curves illustrate the first derivative of the spectra and clearly show a shift of one spectrum towards the blue from the red edge. That spectrum comes from a plant with a concentration of 1377 ppb MITC (Table 2), while the control plant (with the longer wavelength of the maximum inflection point) has a concentration of only 6 ppb MITC.

10.5 DESERTIFICATION/LAND DEGRADATION

An issue of considerable importance in arid and semi-arid environments is that involving land degradation. The stressors bringing about desertification and the response of the ecosystem to these stressors have profound implications in the natural and cultural landscape of areas affected. Methods to assess land degradation are equally diverse, ranging from ground-based soils and vegetation measurements, census and other socioeconomic analyses, and synoptic (including remote sensing) techniques.

Land degradation, or desertification, is a phenomenon involving climate, soils, flora, fauna, and humans. It is the process of change in these ecosystems which can be measured by reduced productivity of desirable plants, alterations in the biomass, and diversity of the micro and macro fauna and flora, accelerated soil deterioration, and increased hazards for human occupancy (Dregne, 1977). It may be regarded as a form of dry land ecosystem degradation because of human use as well as natural factors. The significance of the deterioration derives from its magnitude in the amount of land and numbers of people affected, the rate at which it occurs, and its implications for the future well-being of mankind. Dryland ecosystems under excessive pressure of human use or changes in land use may undergo a loss in productivity resulting in a possible inability to recover (Reining, 1978). It destroys the food-producing capacity of vast tracts of dryland areas in every continent of the world. Many millions of people living in almost 100 countries suffer its effects (UNEP, 1992).

A serious difficulty in examining land degradation in arid regions is their inherent climatic variability. For example, the more extreme arid regions may receive all of their average annual precipitation in one rainfall event (UNEP, 1992). One, two, or several years of below average rainfall may occur in a given region. The identification, monitoring, and combating of land degradation must take this inherent variability into account.

10.6 INDICATORS OF DESERTIFICATION

Changes in indicators of desertification are guides as to whether desertification is becoming more or less of a problem. An "indicator" is defined as a statistic or the presence of a phenomenon judged to carry a specific informative value. An indicator must be diagnostic of an interrelated set of phenomena. Desertification is a dynamic process. Its indicators must, therefore, also be dynamic in order to show the progress of desertification. A number of indicators of desertification are necessary in order to evaluate its effect on the land. Climatic conditions (precipitation, temperature, wind, etc.) are not included among the indicators of desertification even though they may play a considerable role. Indicators of desertification have been grouped into these categories: physical, biological, and social.

10.6.1 PHYSICAL INDICATORS

Physical indicators of land degradation include soil erosion, salinization, depth to ground water, extent and distribution of surface water, numbers and duration of dust and sand storms, presence or absence of soil crusts, amount of soil organic matter, and quality of surface runoff. A potentially valuable physical indicator of desertification is surface reflectance, or albedo. Albedo, or the degree to which light is reflected, can provide integrated information on plant cover and density, soil erosion, salinization, waterlogging, and soil moisture.

Both soil and ground water are subject to salinization, particularly where irrigation has been used on arid and subhumid land. A high evapotranspiration rate in an arid climate can result in the buildup of soil minerals. Even when water used for irrigation is not of particularly poor quality, if the evapotranspiration rate is very high the water evaporates, rather than draining off or leaching through the soil, leaving whatever minerals the water contained and causing a gradual buildup of salts or alkali. Salts and alkali within the soil are also raised to the surface through capillary action. Methods to assess salinization are provided in a separate section within this chapter.

Degradation of soil, involving a number of soil physical and chemical properties, accompanies these changes (these include a decrease in soil permeability, porosity, depth of penetration of water, and amount of available water). At an advanced stage, saline and alkali deposits are visible as white, grey, or black patches. These surface manifestations are clearly visible from airborne and satellite-hosted sensors and, in fact, may be discriminated at an early stage through the use of these types of techniques.

10.6.1.1 Biological indicators

Biological indicators of land degradation include changes in structural, functional, and compositional diversity and impoverishment at multiple levels of biological organization (i.e. genetic, population, community, ecosystem, and regional landscape). Specifically, these include species diversity, productivity, cover (including leaf area index), above-ground biomass, absorbed photosynthetically active radiation, yield, and other measurements.

Schlesinger et al. (1990) have examined desertification through changes in ecosystem function and within the context of spatial and temporal distribution of soil resources relative to vegetation. They have found, through research conducted in the Jornada Experimental Range in southern New Mexico, that when net, long- term desertification of productive grasslands occurs, a relatively uniform distribution of water, nitrogen, and other soil resources is replaced by an increase in their spatial and temporal heterogeneity. This heterogeneity leads to the invasion of grasslands by shrubs. Invasion by shrubs must be considered in part due to the absence of wildfire. In these new plant communities, the soil resources are concentrated under shrubs, while wind and water remove materials from intershrub spaces and transport soil materials to new positions in the landscape.

Changes in vegetation cover (and leaf area index) are also indicative of desertification. Vegetation cover is strongly influenced, in rangelands, by the location of surface water sources. A tendency to overgraze around water sources can itself become a focus of desertification.

10.6.1.2 Social indicators

Social indicators of desertification are those that are related to human occupancy of areas subject to desertification. They include changes in land use and water use (including irrigation, dryland agriculture, and pastoralism), settlement patterns (including diversification of settlement and abandonment), human biological parameters (such as health indices), and social process parameters (such as migration, redistribution, and marginalization). Socioeconomic indicators applicable to people living in areas undergoing desertification may not be uniquely related to desertification. These indicators are general, and are typically used to examine the behaviour of people whose actions may lead to desertification or may be responding to its effects. To the extent that desertification is anthropogenic, i.e., caused by the impact of man on the semi-arid environments, social indicators may also serve as early warning signals (adapted from Reining, 1978).

10.7 MEASUREMENT OF DESERTIFICATION

Several written accounts and photographic records of the western United States illustrate a pattern that is nearly ubiquitous for the region: the vegetation and other land cover of the region existing in the late nineteenth century is considerably different from that which has developed since (Bahre, 1991; Hastings and Turner, 1965; Humphrey, 1956). A quantitative baseline, however, is absent. Various resource inventories may be used to establish such a baseline under present conditions.

The US National Cooperative Soil Survey (NCSS) provides a mechanism to assess certain issues related to desertification. The revised universal soil loss equation, water erosion prediction project, and the wind erosion prediction models have been developed to predict the amount of soil erosion that will occur on a given soil under various climatic environments and management practices. These models use a soil survey database to help pinpoint soil degradation through erosion.

Soil surveys and ecological site inventories provide excellent baseline data for site specific and management unit analyses. The information, however, is more difficult to aggregate for more regional analyses of desertification and climate change processes, and replications of the inventories to detect changes over time are expensive. Statistical grid samples represent one solution, but are still time consuming and costly. The use of remote sensing and other measurement techniques in conjunction with initial baseline inventories represents a strong possibility for desertification assessments and other analyses.

A most prominent manifestation of desertification is the land surface albedo. The albedo (reflectance integrated over the upward hemisphere of directions; sometimes referred to as "brightness") of land surfaces is an indication of degradation. Increasing albedos are thought to indicate erosion, salinization, overgrazing, and other deleterious land surface effects (Mouat et al., 1990). Measurements of albedo may be made through the use of satellite observations. Surface albedo helps determine how much solar energy is absorbed and hence the surface temperature and evapotranspiration.

Various investigators (Justice et al., 1985; Malo and Nicholson, 1990) have made use of remote sensing for deriving vegetation indices for the purpose of assessing vegetation parameters associated with desertification. One such index is the Normalized Difference Vegetation Index (NDVI). The NDVI is typically given as

 where Ch1 represents data from a visible channel (typically a red band -e.g., 0.63-0.69 µm) and Ch2 represents data from a near infrared channel (e.g., 0.76-0.90 µm). Two satellite sensors that are often used for the assessment of NDVI are the Advanced Very High Resolution Radiometer (AVHRR) and the Landsat Thematic Mapper (TM). The AVHRR with its high temporal resolution (images are acquired at every location twice daily) has enjoyed increased use in desertification (through the use of NDVI) studies. NDVI has been used in Africa and elsewhere to demonstrate the effects of overgrazing, especially during times of drought (Justice et al., 1985). The NDVI and other vegetation indices might allow us to obtain an understanding of trends of vegetation phenology in the context of climatic and cultural events. Malo and Nicholson (1990) used AVHRR-derived NDVI to study the dynamics of vegetation and rainfall in the Sahel of West Africa. They found that the temporal and spatial patterns of monthly NDVI closely replicate those of rainfall. They also found that the ratio of NDVI to rainfall provides a rough quantitative measure of the efficiency of water use, with the highest efficiencies found in the plant formations of the driest environments.

While many consider the AVHRR to have too coarse a spatial resolution to be of use for most detailed studies, the Landsat multispectral scanner (MSS) and TM as well as the French SPOT satellite system provide considerably more spatial and spectral detail, and may provide information on both soil salinity and erosion as indicators of desertification.

While remote sensing techniques provide a spatial picture of land processes, other techniques must be employed if the recently acquired (since 1972) satellite images are to be placed into a longer term perspective to assess longer term climate changes. Changes in climate can be examined through the use of the Palmer Drought Severity Index (PDSI). The PDSI is a useful climatic integration that can link modern climatic data to biological processes. It is a useful measure of recent (approximately 100 years) climatic variations, and is derived from a combination of monthly precipitation, temperature, and soil moisture retention information. It offers an integrated measure of moisture availability, i.e., effective precipitation, (Wharton et al., 1990).

If PDSI is combined with other data layers, including those that can be considered to be anthropogenic, an environmental index that perhaps more closely mirrors the pattern of land degradation can be developed. The Desert Research Institute in conjunction with the Environmental Protection Agency (under the auspices of the Environmental Monitoring and Assessment Program, or EMAP) has developed a preliminary Drylands Risk Index to characterize land degradation. The data layers used to develop this index include PDSI, vegetation type, vegetation greenness, total herbivory (measured in terms of total animals vs. carrying capacity), and demographics. Other data layers that could be added include exotics or native species and soil erosion.

The Drylands Risk Index currently provides equal weighting to each of the data layers within the context of a geographic information system. The method of developing ecological risk assessments for the analysis of ecosystem degradation in arid zones has a considerable benefit in allowing scientists and land managers the opportunity to change values of one or more data layers. This could be done to model ecosystem response given either a change in management status or in natural stressors (e.g., climate change).

10.8 REFERENCES

Bahre, C.J. (1991) A Legacy of Change. University of Arizona Press, Tucson, 231 pp. 

Baker, D.E., and Amacher, M.C. (1982) Nickel, copper, zinc, and cadmium. In: Page, A.L. (Ed.) Methods of Soil Analysis, Part 2, 2^nd edn., pp. 323-336. Agronomy Monograph No. 9. American Society of Agronomy, Madison, Wisconsin.

Dawson, J.L., and Nash, T.H. (1980) Effects of air pollution from copper smelters on a desert grassland community. Environ. Exp. Bot. 20, 61.

Dregne, H.E. (1977) Desertification of arid lands. Econ. Geogr. 53, 322-331.

Hastings, J.R., and Turner, R.M. (1965) The Changing Mile. University of Arizona Press, Tucson, 317 pp.

Hill, A.C., Hill, S., Lamb, C., and Barrett, T.W. (1974) Sensitivity of native desert vegetation to SO₂ and NO₂ combined. J. Air Pollut. Control Assoc. 24, 153-157.

Humphrey, R.R. (1956) History of vegetational changes in Arizona. Ariz. Catalog 11, 32-35. 

Keeney, D.R., and Nelson, D.W. (1982) Nitrogen-inorganic forms. In: Page, A.L. (Ed.) Methods of Soil Analysis, Part 2, 2^nd edn., pp. 663-679. Agronomy Monograph No.9. American Society of Agronomy, Madison, Wisconsin.

Justice, C.O., Townshend, R.G., Holben, B.N., and Tucker, C.J. (1985) Analysis of the phenology of global vegetation using meteorological satellite data. Int. J. Remote Sens. 6, 1271-1318.

Kepner, W.G., and Fox, C.A. (Eds.) (1991) Environmental Monitoring and Assessment Program; Strategic Monitoring Plan for Arid Ecosystems. US Environmental Protection Agency, Las Vegas, Nevada.

McAuliffe, J.R. (1988) Markovian dynamics of simple and complex desert plant communities. Am. Nat. 131, 459-490.

MacMahon, J.A. (1981) Introduction. In: Goodall and Perry (Eds.) Arid Land Ecosystems; Structure, Functioning, and Management, Vol. 2, pp. 263-268. International Bioi. Prog. No.17, Cambridge University Press, Cambridge.

Malo, A.R., and Nicholson, S.E. (1990) A study of rainfall and vegetation dynamics in the African Sahel using normalized difference vegetation index. J. Arid Environ. 19, 1-24.

Mouat, D.A., Fox, C.A., and Rose, M.R. (1990) Ecological indicator strategy for monitoring arid ecosystems. In: Proceedings of the International Symposium on Ecological Indicators, 16-19 October 1990, Ft Lauderdale, Florida. Elsevier Science, New York.

Nieto, K.F., and Frankenberger, W.T., Jr. (1985a) Single column ion chromatography: I. Analysis of inorganic anions in soils. J. Soil Sci. Soc. Am. 49, 87-592.

Nieto, K.F., and Frankenberger, W.T., Jr (1985b) Single column ion chromatography. II. Analysis of ammonium, alkali metals and alkali earth cations in soils. J. Soil Sci. Soc. Am. 49, 592-596.

Reining, P. (Ed.) (1978) Handbook on Desertification Indicators. American Association for the Advancement of Science, Washington, D.C., 141 pp.

Rhoades, J.D. (1982) Soluble salts. In: Page, A.L. (Ed.) Methods of Soil Analysis, Part 2, 2^nd edn., pp. 167-179. Agronomy Monograph No.9. American Society of Agronomy, Madison, Wisconsin.

Schlesinger, W.H., Reynolds, J.F., Cunningham, G.L., Huenneke, L.F., Jarrell, W.M., Virginia, R.A., and W.G. Whitford. (1990) Biological feedbacks in global desertification. Science 247, 1043-1048.

Smith, D.M.S., and Morton, S.R. (1990) A framework for the ecology of arid Australia. J. Arid Environ. 18, 255-278.

Speth, G. (1988) Introduction. In: Sears, P.B. (Ed.) Deserts on the March. Island Press, Washington, D.C.

Sposito, C., and Mattigod, S.V. (1979) GEOCHEM; A Computer Program for the Calculation of Chemical Equilibria in Soil Solutions and Other Natural Water Systems. Kearney Foundation of Soil Science, University of California, Riverside.

Thomas, G.W. (1982) Exchangeable cations. In: Page, A.L. (Ed.) Methods of Soil Analysis, Part 2, 2^nd edn., pp. 159-165. Agronomy Monograph No.9. American Society of Agronomy, Madison, Wisconsin.

Thompson, C.R., Olszyk, D.W., Kats, G., Bytnerowicz, A., Dawson, P.J., and Wolf, J.W. (1984) Effects of ozone and sulfur dioxide on annual plants of the Mojave Desert. J. Air Pollut. Control Assoc. 34, 1017-1022.

Turner, R.M. (1990) Long-term vegetation change at a fully protected Sonoran Desert site. Ecology 7, 464-477.

UCAR (1990) Arid Ecosystem Interactions: Lessons from North American Experiences. Draft Report, Office of Interdisciplinary Earth Sciences, Boulder, Colorado.

UNEP (United Nations Environment Program) (1992) World Atlas of Desertification. Edward Arnold, London.

UNFAO (United Nations Food and Agricultural Organization) (1971-1981) FAO/UNESCO Soil Map of the World, 1:5,000,000, Vols. 1-10. UNESCO, Paris.

Webb, R.H., Steiger, J.W., and Turner, R.M. (1987) Dynamics of Mojave Desert shrub assemblages in the Panamint Mountains, California. Ecology 68, 478-490.

Wharton, R.A., Wigand, P. E., Rose, M.R., Reinhardt, R.L., Mouat, D.A., Klieforth, H.E., Ingraham, N.L., Davis, J.O, Fox, C.A., and Ball, J.T. (1990) The North American Great Basin: a sensitive indicator of climatic change. In: Osmond, C.B., Pitelka, L.F., and Hidy, G.M. (Eds.) Plant Biology of the Great Basin, pp. 323-359. Ecological Studies Vol. 80. Springer-Verlag, Berlin, 375 pp.