Natural and Human Dimensions of Land Degradation in Drylands Causes and Consequences 247-257

Chapter 20

Natural and Human Dimensions of Land Degradation in Drylands:Causes and Consequences

James F. Reynolds · Fernando T. Maestre · Paul R. Kemp · D. Mark Stafford-Smith · Eric Lambin

20.1Introduction

Land degradation in drylands, which is referred to asdesertification, is viewed by many as one of the most criti-cally important issues facing many countries (e.g.,Darkoh 1998; Dregne 1996; Kassas 1995). The UnitedNations portrays it as one of the “most important globalchange issues facing mankind” (UNCCD 1994). Landdegradation is a vital societal concern because of its im-pacts on human populations (food security, economics,sustainability, etc.) and environment quality (dust storms,trace gas emissions to the atmosphere, soil erosion, etc.)(Vitousek et al. 1997). Like global climate change and bio-diversity, desertification is the subject of an internationalframework convention, the Convention to Combat De-sertification (CCD), the aim of which is to “target pov-erty, drought and food insecurity in dryland countriesexperiencing desertification, particularly those in Africa”(United Nations 1994). The CCD was established by theUnited Nations in order to facilitate the role of nationalgovernments in enacting policies to combat land degra-dation. The Convention provides for a large infrastruc-ture (e.g., a Secretariat and the Global Mechanism), whichis designed to mobilize and channel financial resources,including the transfer of technology to developing coun-tries (details in Chasek and Corell 2002).

However, in spite of its high profile and acknowledgedimportance, desertification has remained stubbornly in-tractable. Part of the problem is that desertification tendsto stir up more disagreement and controversy than con-sensus (e.g., Leach and Mearns 1996; Reynolds andStafford Smith 2002a; Thomas and Middleton 1994).While the reasons underlying the uncertainty and con-fusion associated with this topic are numerous (see re-views by Grainger et al. 2000; Reynolds 2001; Reynoldsand Stafford Smith 2002a), the bottom line is that deser-tification has proven to be a complex problem that is notamenable to simple solutions. A second aspect of theproblem appears to reside with the CCD itself. The CCDmodel has been roundly criticized for its deficiency ofdirected research efforts, for lacking connections to “real-world” problems, and for serving solely as a mechanismfor some countries to elicit funds from donor nations

(Toulmin 2001). In spite of these criticisms, the CCD is theprimary vehicle for addressing desertification and hasinspired some successful spin-off activities that engagescientists and various stake-holders (see Corell 1999).In this chapter we provide a brief overview of the mostsignificant issues of land degradation in global drylands,with an emphasis on the interaction between human andnatural dimensions of the problem. Despite considerablework on case studies of land degradation in individualregions of the globe, there is little integration betweensocial and biophysical sciences, and there are great op-portunities for comparative studies across the many dif-ferent social and biophysical systems. We discuss someof the underlying causes of land degradation and theirconsequences. Further, we discuss a joint initiative ondesertification of the Global Change and Terrestrial Eco-systems (GCTE) and Land-Use and Land-Cover Change(LUCC) programs, which provided a framework to fa-cilitate directed research effort and progress on this im-portant global environmental issue. Land degradationis an excellent topic for such an initiative given that LUCCis geared to improve the understanding of land-use andland-cover change dynamics, GCTE was focused on syn-thesis activities on critical topics in the terrestrial bio-sphere, and both programs are keen to engage both thephysical and social science communities for the devel-opment of science relevant to global change.

20.2Drylands, Desertification, Drivers, and Scales20.2.1Distribution of People and Land-Cover TypesIn general, drylands are characterized by low and vari-able rainfall, extreme air temperatures, and seasonallyhigh potential evaporation. Technically, drylands aredefined as regions that have an index of aridity (ratio ofmean annual precipitation to mean annual potentialevapotranspiration) of 0.05 to 0.65 (see Middleton andThomas 1997). Drylands can be further subdivided asarid (0.05–0.20 index of aridity; ca. 30% of drylands),semiarid (0.20–0.50 index of aridity, ca. 45% of drylands),and sub-humid (0.50–0.65 index of aridity, ca. 25% ofdrylands) and together cover approximately 5.2 billion

248CHAPTER 20 · Natural and Human Dimensions of Land Degradation in Drylands: Causes and Consequences

hectares or about 40% of the land surface of the globe(Table 20.1a). Drylands have two primary types of hu-man uses: the overwhelming majority serve as rangelands(>75% of drylands), while nearly 20% are rainfed or ir-rigated cropland. In terms of a land-cover classification,shrubland is the dominant cover type (24% of drylands),followed by cropland (20%), savanna (15%), grassland(13%), forest (8%) and urban (3%) (Table 20.1b).

Drylands are home to over two billion people: 42% ofthe Asian population, 41% of Africans, and 25 to 30% ofthe rest of the world (Table 20.1c). Combined, Asia andAfrica contain 84% of all global drylands, dwarfing theamount of dryland area on other continents. In terms ofimportance, however, these numbers can be somewhatmisleading. While Europe contains only ca. 6% of theworld’s drylands, this represents about 32% of its landmass and is home to 25% of its population. Similarly, Aus-

tralia contains about 13% of the world’s drylands but theycover over 75% of the continent and are home to 25% ofits population. Hence in both Europe and Australia, dry-lands are crucial determinants of the economy, cultureand climate. Some of the highest densities of humanpopulations are in the dry sub-humid and semiarid ar-eas of cropland, e.g., those in India, eastern China, andEurope, and in the savannas of Africa (White et al. 2003).

20.2.2Defining Land Degradation

and Desertification

Land degradation and desertification are composite phe-nomena that have no single, readily identifiable attribute.Perhaps this is why there are so many conflicting and

confusing definitions (see reviews by Reynolds 2001;

Thomas 1997). For example, a common misunderstand-ing among land managers and stakeholders is to equateland degradation solely with soil degradation. The defi-nition of Stocking and Murnaghan (2001) emphasizeschanges in biophysical variables and implies an impacton human populations: land degradation is how one ormore land resources (soils, vegetation, water, landforms,etc.) has changed “for the worse,” signifying a temporaryor permanent decline in the productive capacity of theland. While heuristic, we favor the definitions used bythe CCD (UN 1994), which go a step further by making itclear that while biophysical components of ecosystemsand their properties are involved, the interpretation ofchange as ‘loss’ is dependent upon the integration of thesecomponents within the context of the socio-economic ac-tivities of human beings. The CCD’s definition states thatland degradation is the reduction or loss of the biologi-cal and economic productivity and complexity of terres-trial ecosystems, including soils, vegetation, other biota,and the ecological, biogeochemical, and hydrologicalprocesses that operate therein. In drylands, this involvessoil erosion and sedimentation, shifts in natural firecycles, the disruption of biogeochemical cycling, and areduction of native perennial plants and associated mi-crobial and animal populations. The CCD’s definition ofdesertification explicitly focuses on the linkages betweenhumans and their environments that affect human wel-fare in arid and semi-arid regions.

20.2.3What Drives Land Degradation

and Desertification?

Desertification is caused by a relatively large number offactors that vary from region to region, and that oftenact in concert with one another in varying degrees. Geistand Lambin (2004) carried out a worldwide review ofthe causes of desertification, and from 132 case studiesidentified four major categories of proximal causalagents: (1) increased aridity; (2) agricultural impacts,including livestock production and crop production;(3) wood extraction, and other economic plant removal;and (4) infrastructure extension, which could be sepa-rated into irrigation, roads, settlements, and extractiveindustry (e.g., mining, oil, gas). They concluded that onlyabout 10% of the case studies were driven by a singlecause (with about 5% due to increased aridity and 5% toagricultural impacts). About 30% of the case studies wereattributable to a combination of two causes (primarilyincreased aridity and agricultural impacts), while theremaining cases were combinations of three or all fourproximal causal factors.

A primary objective of Geist and Lambin’s (2004) re-view was to identify general global patterns in causa-tion of desertification. As such, the study identified spe-cific agents as more or less important in particular re-

20.2 · Drylands, Desertification, Drivers, and Scales249

gions, and indicated that these agents derive from un-derlying forces associated with particular combinationsof socio-economic (including technology) and biophysi-cal factors characteristic of particular regions. For ex-ample, two underlying forces, climate and technologi-cal factors (either new technologies or deficiencies intechnology) were the key drivers of desertification inthe majority (54%) of the case studies in southern Eu-rope. In Africa, climate, alone or acting in concert withpopulation demography, was a key driver in the 38% ofthe case studies. In the United States, 50% of the casestudies were attributable to a combination of climateand technology drivers or these two factors interactingwith economic forces. Desertification processes in Asia,Latin America, and Australia could only be attributedto more complex interactions among four or more un-derlying forces.

20.2.4Estimating the Extent of DesertificationNot surprisingly, obtaining accurate estimates of theamount of drylands affected by desertification is a dif-ficult task, fraught with numerous obstacles and com-plications. Nevertheless, the extent of global desertifi-cation is routinely reported as high as 70% of all dry-lands (UNCCD 2000)! Thomas and Middleton (1994)and others view such estimates as suspect because theyare largely derived from the subjective judgments ofscientists and laypersons, surveys completed by localgovernments, qualitative assessments, and data of vary-ing authenticity and consistency. Unfortunately, the CCDdefinition of desertification (Sect. 20.2.2), which we con-sider most authoritative at present, is not amenable toeasy quantification, especially as a single number or syn-thetic index.

A variety of problems confound estimates of the ex-tent of desertification. For instance, observations madeon short-term ecosystem dynamics are often cited asevidence of desertification ignoring the fact that drylandsare highly variable over time (both intra- and inter-an-nually) and that a temporary loss of vegetation cover dueto a short-term drought or to local land use (e.g., graz-ing) is distinct from – and not necessarily related to – apermanent loss of vegetation associated with desertifi-cation (Reynolds 2001). Inaccurate estimations are alsofueled by technological barriers that preclude the moni-toring of relevant variables, such as the cover of dry veg-etation, that cannot be easily estimated using traditionalapproaches (Asner et al. 2003). Another crucial, but of-ten overlooked, concern is that desertification is usuallypromoted by two or more causal agents (see Sect. 20.2.3).In spite of the CCD definition, most estimates of deserti-fication are derived solely from either biophysical fac-tors (e.g., soil erosion, loss of plant cover, change in al-bedo) or socio-economic factors (decreased production,

250CHAPTER 20 · Natural and Human Dimensions of Land Degradation in Drylands: Causes and Consequences

economic loss, population movements, etc.), but rarelyboth types simultaneously (Stafford Smith and Reynolds2002). When assessments are made without good knowl-edge of the underlying causes, it brings into question thevalidity of the variables or sets of variables being used inthe assessment.

Over the years, in different arid and semiarid regionsof the world, there has been a concerted effort tocategorize and map various forms of land degradationat various scales, but these efforts have failed to in-clude a careful, systematic identification of the criticalvariables that cause the observed dynamics (StaffordSmith and Reynolds 2002). This problem lies at thebase of the confusion about how much ‘desertification’there really is (see Batterbury et al. 2002). StaffordSmith and Reynolds (2002) argued that much of thisconfusion could be eliminated by focusing on a smallnumber of critical variables that contribute to an un-derstanding of the cause, rather than effect, of desertifi-cation. Of course, this is all the more problematic whenwe try to account for the causal factors driving deserti-fication in different regions of the world and at differ-ent times: approaches developed to estimate desertifi-cation in one region may not be effective in others. Thefailure to recognize these issues has led to the dispari-ties of estimates of desertification in the literature andis responsible for many of the disagreements alluded toabove (Stafford Smith and Pickup 1993; Stafford Smithand Reynolds 2002).

20.2.5Consequences of Desertification

There are few disagreements that desertification has alarge number of biophysical and socio-economic conse-quences, which range across a wide spectrum of spatialand temporal scales. An in-depth treatment of the dif-ferent consequences is beyond the scope of this chapter,but some relevant ones are presented in Table 20.2 (andsee group reports in Reynolds and Stafford Smith 2002b).From the socio-economic point of view, most conse-quences (especially in pastoral systems) are a direct con-sequence of the decline in ‘productivity’ or the capacityof the land to support plant growth and animal produc-tion. During early stages of desertification such lossesare compensated by the social resilience of the local hu-man populations, especially in developing countries, orby economic inputs from government (Vogel and Smith2002). However, when certain thresholds are crossed, so-cial resilience or government subsidies may not be enoughto compensate for the loss of productivity, and this fuels abattery of socio-economic changes that range from modi-fications in trade promoted by lower agricultural produc-tion to large population migrations (Fernández et al. 2002).Virtually all of the biophysical consequences start withthe loss of vegetation and soil (Table 20.2). These losseshave a ‘cascading’ effect on other components and processes,leading to a progressive deterioration of the ecological

structure and functioning of the system (Fernández et al.

20.2 · Drylands, Desertification, Drivers, and Scales251

2002). The specific biophysical consequences of deserti-fication differ substantially between geographical areasof the globe as a function of the intensity and number ofdriving forces at work, the extent of the impacted area,the duration of the deterioration, and the resilience ofthe system components (especially vegetation). Evenwithin a particular area we may find that there are dif-ferent consequences depending upon the unique char-acteristics of the system. For instance, while Krogh et al.(2002) found that some keystone species associated withgrasslands are negatively affected by shrub encroach-ment into former grasslands (a form of desertification,Schlesinger et al. 1990), recent studies in the southwest-ern United States have shown that shrub encroachmentis associated with an increase in the species richness ofbirds (Pidgeon et al. 2001), mammals (Whitford 1997)and ants (Bestelmeyer 2005). Thus, some generalizationsregarding biophysical consequences of desertificationmay be misleading or incorrect when applied to spe-cific situations.

20.2.6Scale and Hierarchy

The importance of scale is manifest in a number of wayswhen evaluating the extent and effects of desertifica-tion (Sect. 20.2.4). Obviously, humans are most con-cerned with the local subset of degradation that impacts

them personally. Reynolds and Stafford Smith (2002a)use a hypothetical case-study of gully formation (over-grazing by cattle leads to loss of vegetative cover and,ultimately, soil erosion) to illustrate how different seg-ments of society (the stake-holders) see such problemswith differing degrees of concern. Whereas an ecolo-gist might view erosion gullies as a breakdown in eco-system function, this will resonate with a farmer only ifthe gullies have a demonstrable impact on his values,i.e., meat production by his cattle. If not, the farmer willnot consider this as ‘degradation’. Alternatively, otherstake-holders, such as urban dwellers in a nearby town,may consider this localized soil erosion and gully for-mation a more serious problem because of the poten-tial for silt runoff into the town’s reservoir and its ad-verse effects on water quality.

Because local activities often have regional conse-quences (e.g., localized erosion gullies impacting regionalwater quality) and regional issues can, in turn, have localimpacts, it behooves us to assume a multifactor, multi-scale, hierarchical view of desertification. In Table 20.3,we illustrate representative scales of interest in desertifi-cation viewed from both socio-economic and biophysi-cal perspectives. Coupled socio-economic and biophysi-cal systems must be hierarchically nested in order toavoid errors that will undoubtedly occur if we attempt toextrapolate understanding over a range of scales that is

too great, e.g., trying to predict what will happen at the

252CHAPTER 20 · Natural and Human Dimensions of Land Degradation in Drylands: Causes and Consequences

household level based on observations made at the na-tional scale. Sørbø (2003) describes examples from eastAfrica to show the importance of scale in pastoral herd-ing communities. These pastoral communities are net-worked into various localized units, which function in acomplex interplay of local and regional social, economic,and political factors, all of which have evolved over manyyears against a backdrop of severe environmental insta-bility and unpredictable contingencies. It is not surpris-ing that ‘top-down’ attempts to ‘manage’ these systemsfall short of expectations. This is consistent with Batter-bury et al.’s (2002) observations that, in arid and semi-arid lands, the highest levels of the ‘management hierar-chy’ are invariably quite remote from marginal lands and,consequently, have weak political and economic feed-backs. An important objective of institutions such as theCCD should be to provide a context within which levelsin the hierarchy become more integrated and aware ofthe issues (Batterbury et al. 2002; Lambin et al. 2002).

20.3Joint GCTE-LUCC Desertification InitiativeThe simultaneous assessment of biophysical and socio-economic drivers (and consequences) of desertification hasbeen recognized as one of the most challenging – but po-tentially rewarding – topics for further research (see re-view by Reynolds 2001). In an attempt to address thischallenge, the GCTE and LUCC programs of the Inter-national Geosphere-Biosphere Programme joined forcesto establish an initiative on desertification. The intent ofthis initiative was to bring together researchers from thevarious global change programs, representing both natu-ral and human-influenced systems, with the objective ofstimulating, developing, and refining a new paradigm tobear on this important global change concern.

One of the key, initial products of this GCTE-LUCCinitiative was a book on global desertification (GlobalDesertification. Do Humans Cause Deserts?), which ex-plicitly addresses many of the significant interactions andfeedbacks between natural and human-influenced dry-land systems. Focusing on the multitude of interrelation-ships within coupled human-environment systems thatcause desertification, and drawing heavily from the chap-ters of this book, Stafford Smith and Reynolds (2002)proposed the Dahlem Desertification Paradigm (DDP).The DDP is a new synthetic framework that is unique intwo ways:

First, the DDP attempts to capture the multitude ofinterrelationships within human-environment sys-tems that cause desertification, within a single, syn-thetic framework; and

Second, the DDP is testable, which ensures that it canbe revised and improved upon as a dynamic frame-work.

In this section we briefly present an outline of the keyelements or assertions of the DDP and describe onemethod we are using for testing it. As is the case for manyparadigms, the constituent ideas contained within theDDP themselves are generally not new, but rather, theybring together much of the previous work on this diffi-cult topic in a way that reveals new insights.

20.3.1Dahlem Desertification Paradigm

The DDP consists of nine assertions (Table 20.4), whichembrace a hierarchical view of land degradation andhighlight key linkages between socio-economic and bio-physical systems at different scales. The first three asser-tions relate to the working framework of the DDP whilethe remaining ones focus on its implementation, limita-tions, and potentials. The main points of the DDP are:1.that an integrated approach, which simultaneouslyconsiders both biophysical and socio-economic at-tributes in these systems, is absolutely essential tounderstand land degradation (assertions #1, #7);2.that the biophysical and socio-economic attributesthat govern or cause land degradation in any particu-lar region are invariably ‘slow’ (e.g., soil nutrients) rela-tive to those that are of immediate concern to humanwelfare – the ‘fast’ variables (e.g., crop yields). It is nec-essary to distinguish these in order to identify the causesof land degradation from its effects (assertion #2);3.that socio-ecological systems in drylands of the worldare not static (assertions #3, #6);

4.that while change is inevitable, there does exist a con-strained set of ways in which these socio-ecologicalsystems function, thereby allowing us to understandand manage them (assertion #9);

5.that restoring degraded socio-ecological systems tomore productive, sustainable states requires outsideintervention (assertion #4)

6.that socio-ecological systems in drylands of the worldare hierarchical (assertion #8). Hence, scale-relatedconcerns abound and desertification itself is a region-ally-emergent property of localized degradation (as-sertion #5)The strength of the Dahlem Desertification Paradigmis in its cross-scale conceptual holism. While the termdesertification is only useful at the higher levels of ag-gregation, and degradation (appropriately refined) at thelower levels, the DDP framework embraces all these lev-els of concern. For example, at the international level,implementation of the CCD must be framed in terms ofchanges in coupled human-environment systems thatmatter to humans, which dramatically changes the mean-ing of the “extent of desertification” (Sect. 20.2.4) andboth the timing and distribution of funding for inter-

20.3 · Joint GCTE-LUCC Desertification Initiative253

vention. Similarly, at the household or community level,where concern is on the specific type of land degrada-tion that is occurring and its local socio-economic con-sequences, the DDP channels resources towards identi-fying those essential biophysical and socio-economicslow variables that really matter in terms of quantifyingcurrent and future risk.

20.3.2Initiatives to Test the Dahlem

Desertification Paradigm

The joint GCTE-LUCC initiative on desertification isembodied within the ARIDnet1 research network(Reynolds et al. 2003). The general objectives of this net-work are to foster international cooperation, discussion,and exchange of ideas about global desertification (assummarized in the DDP), to conduct case studies, repre-senting a range of biophysical/socio-economic land deg-radation types around the world, and to facilitate com-munication between researchers to foster more practi-cal, field-level interactions with stakeholders in sustain-able land management.

To accomplish these objectives, ARIDnet is organizedinto three nodes (North/South America; Asia/Australia;Europe/Africa) and is pursuing four specific tasks(Fig. 20.1):

Fig. 20.1. ARIDnet is an international research network organizedinto three nodes (North/South America; Asia/Australia; Europe/Af-rica) and is pursuing four tasks (see text): paradigm-building, con-ducting case studies, developing a synthesis, and to facilitate par-

ticipation via network-building

1

Assessment, Research, and Integration of Desertification.

Paradigm-building. Using workshops and symposiaconducted in different parts of the world, ARIDnetwill develop and refine the contents of the DDPthrough the joint participation of the internationalcommunity of desertification researchers, stakehold-ers, and policy-makers;

Case studies. Working Groups are specifically formedto develop case studies throughout the world in orderto test the DDP in a well-stratified, comparative man-ner. These case studies are based on existing data and

specific stakeholders, and represent a wide range of

254CHAPTER 20 · Natural and Human Dimensions of Land Degradation in Drylands: Causes and Consequences

the biophysical and socio-economic conditions exist-ing in drylands;

Synthesis. The numerous case studies will feed into aquantitative assessment of what really matters in de-sertification. This synthesis will especially focus onthose interactions between key biophysical and socio-economic variables; and

Network-building. ARIDnet strives to recruit and fos-ter the participation of a diversity of researchers fromdifferent fields and countries in the activities of thenetwork.

20.4Management of Desertified Drylands

In preceding sections we provided examples of the magni-tude and importance of desertification as a major local andglobal environmental problem. In order to prevent furtherdegradation and to restore degraded lands, a number ofcountries have enacted environmental policies to estab-lish management actions to combat desertification. Theseactions are diverse but can be grouped under three head-ings: avoidance, monitoring, and restoration.

20.4.1Avoidance

Management actions to avoid desertification are rarelyproactive. If they exist, they normally are focused on thehuman drivers of desertification. Such actions varywidely according to the socio-economic conditions of aparticular country, but often include the use of economicsubsidies to promote changes in land use or crops (Harou2002), the diversification of human activities in the ar-eas affected (Pamo 1998), and the establishment of edu-cational programs to improve education and social wel-fare (Vogel and Smith 2002). The latter activity is of cru-cial importance since one of the core causes of desertifi-cation in developing nations is the extreme pressure onthe land resulting from high population growth (Geistand Lambin 2004; Le Houérou 1996). However, there areexamples of success in reducing demographic growth andin implementing sustainable production systems in de-sertified areas worldwide (Arkutu 1995; Vogel and Smith2002), indicating that with appropriate resources andpolitical willingness, some of the most important deser-tification drivers can be controlled.

20.4.2Monitoring

The monitoring of desertification is an increasingly im-portant development in the management of dryland ar-eas. The establishment of long-term and rigorous moni-toring programs is an effective way to assess the statusof natural resources and the evolution of desertification

processes. Such programs could provide an “early warn-ing” of pending concerns, e.g., the detection of changesin ecosystem attributes and processes at stages wheremanagement actions would be most cost-effective (seeFernández et al. 2002).

It is encouraging to see increased research dedicatedto the development of easily accessible monitoring meth-ods based on simple soil and vegetation indicators (e.g.,Pyke et al. 2002; Tongway and Hindley 1995). These meth-ods are based on the collection of basic information ofthose vegetation attributes and soil properties (e.g., cover,spatial pattern, resistance to penetration and texture) thatlargely determine the ecosystem’s resilience to erosiveforces and its ability to use water and nutrients. An im-portant goal of these methods is to minimize the train-ing and equipment required so as to increase their avail-ability to nations with low economic resources. Theseground-based methods are complemented with the useof remote sensing data, which have been successfullyemployed to monitor desertification processes in the U.S.(Asner and Heidebrecht 2005), South America (Asner andHeidebrecht 2003), Africa (Prince et al. 1998), Australia(Bastin et al. 2002) and Europe (Imeson and Prinsen2004). Remote-sensing approaches are often based onmeasuring the same vegetation and soil attributes asground-based methods, but they allow the establishmentof monitoring programs at larger spatial scales. How-ever, they often require the use of expensive equipmentand appropriate training, two factors not available inmany countries. Furthermore, recent studies suggest thattraditional remote sensing measurements (e.g., NDVImeasured during the growing season) do not provide anadequate indication of biophysical degradation (Asneret al. 2003). Further advancement in desertificationmonitoring depends upon a coordination betweenground- and remote sensing-based research and the es-tablishment of sound and cost-effective methodologiesappropriate for particular regions.

20.4.3Restoration

While we might hope that future actions of land owners,communities, regions, and nations will begin to adoptmanagement practices and policies that minimize oravoid desertification, the sad fact is that vast areas of dry-lands are already in a wretched state, with varying de-grees of reduced plant cover, impoverished species di-versity, and depleted or eroded soils (Whisenant 1999).Restoration actions in these lands have traditionally fo-cused on the biophysical variables, especially those aimedat increasing plant cover and halting soil erosion. Often,and irrespective of the underlying drivers of desertifica-tion, restoration efforts involve the establishment ofwoody plants (Whisenant 1999), which is deemed cru-cial to stop further degradation, and to foster recovery

of ecosystem structure, composition, and functioning(Reynolds 2001). Such afforestation programs, whichhave been carried out since the 1800s, have resulted inmillions of hectares of conifer trees (mainly) planted indrylands of Spain, Turkey, Morrocco, Algeria, Argentina,China, and many other countries (Richardson 1998;Pausas et al. 2004; Sánchez Martínez and Gallego Simón1993; Gao et al. 2001). While some of these efforts havebeen effective in controlling desertification, many havefailed (Odera 1996), and in some cases have exacerbatedthe degradation process (see review by Maestre andCortina 2004).

In other instances, restoring desertified lands has in-volved instituting changes in land management practices(e.g., Pyke et al. 2002; Sørbø, 2003). Again, these may beundertaken without full understanding of the causalmechanisms involved, or without an appreciation of thesocio-economic conditions (Sørbø, 2003), and again areprone to failure. This history of lack of success suggestsa clear need to address fundamentals. First, as we haveemphasized throughout this chapter, it is important tohave a sound understanding of both the biophysical andhuman dimensions of causality. Second, we must incor-porate as much of our knowledge of dryland structureand functioning as possible into our management ac-tions. In this respect, important advances have been madein the application of conceptual and practical ecosystemmodels. Third, we must carefully evaluate and incorpo-rate existing socio-economic attributes. Fourth, we mustassess the potential for successful rehabilitation. Lastly,we must strive for improved restoration methodologies,including the use of new understanding of key ecosys-tem processes, e.g., plant-plant interactions (Maestre et al.2001) and the role of soil heterogeneity in plant estab-lishment (Maestre et al. 2003).

20.5Summary and Conclusions

While conceptual, methodological and technologicaladvances have been made during the past several decadesto help land managers establish appropriate strategiesfor combating desertification, there have been relativelyfew successes in desertification abatement (Le Houérou1996). As discussed in this chapter, desertification mayhave numerous underlying causes, which involve a com-plex interplay among biophysical and human dimensions.We propose that a key aspect of the prevention and re-mediation of desertification is the development of a in-tegrative, theoretical framework in which biophysical andsocioeconomic components are treated as coupled pro-cesses fostering desertification.

We further suggest that the Dahlem DesertificationParadigm provides the necessary theoretical frameworkfor bringing researchers and policy-makers together forthe purpose of developing testable hypotheses and meth-References255

odologies regarding the monitoring, prevention and re-mediation of desertification. The DDP emphasizes theneed to address all socio-economic levels (local, regional,national, international) in the development of effectivedesertification policy decisions. The DDP emphasizesmoving beyond isolated studies of various parts of thedesertification problem and toward establishing an in-tegrated program of causal links of dryland degradation,from climate dynamics to ecological impacts to policyresponse strategies, which can be applied to a wide rangeof temporal and spatial scales. The challenges of build-ing the necessary political and scientific bridges are enor-mous, but so is the need for urgent action to understandand manage desertified drylands worldwide.

Acknowledgments

The authors acknowledge the helpful comments and cri-tiques provided by Greg Asner and an anonymous re-viewer. This research was supported by the National Sci-ence Foundation (grant NSF-DEB-02–34186 to ARIDnet)and the Center for Integrated Study of the Human Di-mensions of Global Change, through a cooperative agree-ment between the NSF (SBR-9521914) and CarnegieMellon University. FTM acknowledges support via aFulbright fellowship from the Spanish Ministry of Edu-cation and Science, funded by Secretaría de Estado deUniversidades and Fondo Social Europeo. JFR acknowl-edges partial support by the Alexander von Humboldt-Stiftung during the preparation of this manuscript andthe excellent hospitality of the Lehrstuhl für Pflanzen-ökologie at the University of Bayreuth.

References

Arkutu A (1995) Family planning in Sub-Saharan Africa: PresentAsner GPstatus and future strategies. Int J Gyn Obst 50: S27–S34

tification studies: Comparing A, Heidebrecht KB (2003) Imaging spectroscopy for deser-Asner GP, Heidebrecht KB (2005) Desertification alters regionalArgentina drylands. IEEE Trans Geosci Rem Sens 41:1283–1296VIRIS and EO-1 Hyperion inAsner GPecosystem-climate interactions. Global Change Biol 11:182–194Argentina: Changes in ecosystem carbon and nitrogen from, Borghi CE, Ojeda RA (2003) Desertification in centralAsner GPimaging spectroscopy. Ecol Appl 13:629–648

ing systems, ecosystem responses, and global change. Annu Rev, Elmore AJ, Olander LP, Martin RE, Harrias T (2004) Graz-Bastin G, Ludwig JA, Eager RWEnviron Resour 29:261–299

, Chewings VH, Liedloff AC (2002)Indicators of landscape function: comparing patchiness metricsusing remotely-sensed data from rangelands. Ecol Indic 1:243Batterbury SPJ, Behnke RH, Döll PM, Ellis JE, Harou PA, Lynam TJP,Mtimet A, Nicholson SE, Obando JA, Thornes JB (2002) Respond-ing to desertification at the national scale: detection, explana-tion, and responses. In: Reynolds JF, Stafford Smith M (eds) Glo-bal Desertification: Do Humans Cause Deserts? Dahlem Uni-versity Press, Berlin, pp 357–386

Belnap J, Eldridge DJ (2001) Disturbance and recovery of biological soilcrusts. In: Belnap J, Lange OL (eds) Biological Soil Crusts: Structure,Function, and Management Springer-Verlag, Berlin, pp 363–384

256CHAPTER 20 · Natural and Human Dimensions of Land Degradation in Drylands: Causes and Consequences

Bestelmeyer BT (2005) Does desertification diminish biodiversity?Enhancement of ant diversity by shrub invasion in south-west-Bollig M, Schulte A (1999) Environmental change and pastoral per-ern USA. Divers Distrib 11:45–55

ceptions: Degradation and indigenous knowledge in two Afri-Chasek Pcan pastoral communities. Human Ecol 27:493–514

tional level: the institutional system. In: Reynolds JF, Stafford, Corell E (2002) Addressing desertification at the interna-Smith M (eds) Do Humans Create Deserts? The Ecological,Meteorological, and Human Dimensions of Global Desertifica-Corell E (1999) The Negotiable Desert: Expert Knowledge in thetion. Dahlem University Press, Berlin, pp 275–296

Negotiations of the Convention to Combat Desertification.Linköping Studies in Arts and Science No. 191. Linköping Uni-Darkoh MBK (1998) The nature, causes and consequences of de-versity, Linköping, Sweden

sertification in the drylands of Africa. Land Degrad DevelopDregne HE (1996) Desertification: challenges ahead. Ann Arid Zone9:1–20

Fernández RJ, Archer ERM, Ash AJ, Dowlatabadi H, Hiernaux PHY,35:305–311

Reynolds JF, Vtion and recovery in socio-ecological systems: a view from theogel CH, Walker BH, Wiegand T (2002) Degrada-household/farm level. In: Reynolds JF, Stafford Smith M (eds)Do Humans Create Deserts? The Ecological, Meteorological, andHuman Dimensions of Global Desertification. Dahlem Univer-Fredrickson E, Havstad KM, Estell R, Hyder P (1998) Perspectivessity Press, Berlin, pp 297–324

on desertification: South-western United States. J Arid EnvironGallart F, Llorens P39:191–207

tural terraces on runoff generation in a small Mediterranean, Latron J (1994) Studying the role of old agricul-Gao J, Zha Y, Ni S (2001) Assessment of the effectiveness of deserti-mountainous basin. J Hydrol 159:291–303

fication rehabilitation measures in Yulin, north-western ChinaGeist HJ, Lambin EF (2004) Dynamic causal patterns of desertifi-using remote sensing. Int J Rem Sens 22:3783–3795

Gonzalez P (2001) Desertification and a shift of forest species in thecation. Bioscience 54:817–829

Grainger A, Stafford Smith M, Squires VWest African Sahel. Clim Res 17:217–228

cation and climate change: the case for greater convergence., Glenn E (2000) Desertifi-Harou PA (2002) What is the role of markets in altering the sensi-Mitig Adapt Strat Global Change 5:361–377

tivity of arid land systems to perturbation? In: Reynolds JF,Stafford Smith DM (eds) Global Desertification: Do HumansHuenneke LF, Anderson JP, Remmenga M, Schlesinger WH (2002)Cause Deserts? Dahlem University Press, Berlin, pp 253–274Desertification alters patterns of aboveground net primaryproduction in Chihuahuan ecosystems. Global Change BiolImeson AC, Prinsen HAM (2004) V8:247–264

indicators for identifying runoff and sediment source and sinkegetation patterns as biologicalareas for semi-arid landscapes in Spain. Agric Ecosyst EnvironJackson RB, Banner JL, Jobbagy EG, Pockman WT, Wall DH (2002)104:333–342

Ecosystem carbon loss with woody plant invasion of grasslands.Kassas M (1995) Desertification: a general review. J Arid EnvironNature 418:623–626

Kelley DW30:115–128

watersheds into Lake Pepin, Minnesota, USA. J Environ Qual, Nater EA (2000) Historical sediment flux from threeKrogh SN, Zeisset MS, Jackson E, Whitford WG (2002) Presence/29:561–568

absence of a keystone species as an indicator of rangeland mbin EF, Chasek PS, Dowing TE, Kerven C, Kleidon A, LeemansJ Arid Environ 50:513–519

R, Lüdeke M, Prince SD, Xue Y (2002) The interplay betweeninternational and local processes affecting desertification. In:Reynolds JF, Stafford Smith M (eds) Do Humans Create Deserts?The Ecological, Meteorological, and Human Dimensions of Glo-Latchininsky A, Gapparov FA (1996) Consequences of the Aral sea’sbal Desertification. Dahlem University Press, Berlin, pp 387–402drying up on the regional locust situation. Secheresse 7:109–113

Lavee H, Imeson AC, Sarah P (1998) The impact of climate changeon geomorphology and desertification along a mediterranean-Le Houérou HN (1996) Climate change, drought and desertifica-arid transect. Land Degrad Develop 9:407–422

Leach M, Mearns R (eds) (1996) The Lie of the Land: Challengingtion. J Arid Environ 34:133–185

Received Wisdom on the African Environment. HeinemannMaestre FT, Cortina J (2004) Are Press, Portsmouth, N.H

ful as a restoration tool in semiarid Mediterranean areas? ForPinus halepensis plantations use-Maestre FT, Bautista S, Cortina J, Bellot J (2001) Potential for usingEcol Manage 198:303–317

facilitation by grasses to establish shrubs on a semiarid degradedMaestre FT, Cortina J, Bautista S, Bellot J, Vsteppe. Ecol Appl 11:1641–1655

scale environmental heterogeneity and spatiotemporal dynam-allejo R (2003) Small-ics of seedling establishment in a semiarid degraded ecosystem.Middleton NJ, Thomas DSG (eds) (1997) WEcosystems 6:630–643

orld Atlas of Desertifica-Odera J (1996) Ltion. U.N. Environment Programmem, Edward Arnold, New Yorkdans les zones sèches et le rôle de la foresterie dans la restauration,’état de dégradation actuel des écosystèmes fragilesNote de secrétariat nº 1. Réunion Internationale d’experts sur larestauration des ecosystèmes forestiers dégradés. PortuguesePamo ET (1998) Herders and wildgame behaviour as a strategy againstMinistry of Agriculture, UNP/UNSO, IPF, CCD & RIOD, LisbonPausas JG, Bdesertification in northern Cameroon. J Arid Environ 39:179–190Vilagrosa A, Bautista S, Cortina J, Vladé C, Valdecantos A, Seva J, Fuentes D, Alloza J,oaks in the restoration of Mediterranean landscapes of Spain:allejo R (2004) Pines andPidgeon AM, Mathews NE, Benoit R, Nordheim EV (2001) ResponseNew perspectives for an old practice. Plant Ecol 171:209–220of avian communities to historic habitat change in the north-Prince SD (2002) Spatial and temporal scales for detection of de-ern Chihuahuan Desert. Cons Biol 15:1772–1788

sertification. In: Reynolds JF, Stafford Smith M (eds) GlobalDesertification: Do Humans Cause Deserts? Dahlem UniversityPrince SD, de Colstoun EB, Kravitz LL (1998) Evidence from rain-Press, Berlin, pp 23–40

use efficiencies does not indicate extensive Sahelian desertifi-Pyke DA, Herrick JE, Shaver PL, Pellant M (2002) Rangeland healthcation. Global Change Biol 4:359–374

attributes and indicators for qualitative assessment. J RangeReynolds JF (2001) Desertification. In: Levin S (ed) EncyclopediaManage 55:584–597

of Biodiversity, Volume 2. Academic Press, San Diego, CA,Reynolds JF, Stafford Smith M (2002a) Do humans cause deserts?pp 61–78

In: Reynolds JF, Stafford Smith M (eds) Global Desertification:Do Humans Cause Deserts? Dahlem University Press, BReynolds JF, Stafford Smith M (eds) (2002b) Global Desertification:pp 1–22

erlin,Reynolds JF, Stafford Smith DM, Lambin EF (2003) ARIDnet: Seek-Do Humans Cause Deserts? Dahlem University Press, Berlining novel approaches to desertification and land degradation.IGBP Global Change Newsletter 54, 5–9 (PDF available at: http:/Richardson DM (1998) Ecology and biogeography of /www.igbp.kva.se)

Pinus. Cam-Rosenfeld D, Rudich Y, Lahav R (2001) Desert dust suppressing pre-bridge University Press, Cambridge

cipitation: A possible desertification feedback loop. Proc NatlSánchez Martínez JD, Gallego Simón VJ (1993) La Política deAcad Sci USA 98:5975–5980

repoblación forestal en España, siglos XIX y XX: Planteamientos,Actuaciones y resultados, estado de la cuestión y recopilaciónbibliográfica. MAPA. Secretaria General Técnica. Serie Reco-Schlesinger WH, Reynolds JF, Cunningham GL, Huenneke LF, Jarrellpilaciones Bibliográficas n° 10, Madrid

WM, Virginia R, Whitford WG (1990) Biological feedbacks inSchlesinger WH, Abrahams AD, Parsons AJ, Wainwright J (1999)global desertification. Science 247:1043–1048

Nutrient losses in runoff from grassland and shrubland habi-tats in Southern New Mexico: I. rainfall simulation experiments.Biogeochemistry 45:21–34

Sharma KD (1998) The hydrological indicators of desertification. JSørbø GM (2003) Pastoral ecosystems and the issue of scale. AmbioArid Environ 39:121–132

Stafford Smith DM, Pickup G (1993) Out of Africa, looking in:32:113–117

understanding vegetation change. In: Behnke RH Jr., ScoonesI, Kerven C (eds) Range Ecology at Disequilibrium: NewModels of Natural Variability and Pastoral Adaptation in Afri-can Savannas. Overseas Development Institute and Interna-tional Institute for Environment and Development, London,Stafford Smith DM, Reynolds JF (2002) The Dahlem Desertifica-pp 196–244

tion Paradigm: A new approach to an old problem. In: ReynoldsJF, Stafford Smith M (eds) Global Desertification: Do HumansStocking M, Murnaghan N (2001) Handbook for the Field Assess-Cause Deserts? Dahlem University Press, Berlin, pp 403–424Thomas DSG (1997) Science and the desertification debate. J Aridment of Land Degradation. Earthscan, London; Sterling, VAThomas DSG, Middleton N (1994) Desertification: Exploding theEnviron 37:599–608

Tongway DJ, Hindley N (1995) Assessment of soil condition of tropi-Myth. John Wiley & Sons, Chichester

Tongway DJ, Ludwig JA (1996) Rehabilitation of semiarid landscapescal grasslands. CSIRO Ecology and Wildlife, Canberra

in Australia. I. Restoring productive soil patches. Restor EcolToulmin C (2001) Lessons from the theatre: Should this be the final4:388–397

curtain call for the Convention to Combat Desertification? WSSDOpinion Series. International Institute for Environment andUNCCD (1994) United Nations Convention to Combat Desertifi-Development. Available from:

cation, Intergovernmental Negotiating Committee For a Con-vention to Combat Desertification, Elaboration of an Interna-tional Convention to Combat Desertification in Countries Ex-periencing Serious Drought and/or Desertification, Particu-larly in Africa. U.N. Doc. A/AC.241/27, 33 I.L.M. 1328. UnitedNations, New York

References

257

UNCCD (2000) Fact Sheet 2: The Causes of Desertification. UnitedNations Secretariat of the Convention to Combat Desertifica-tion. Available from: http://www.unccd.int/publicinfo/factsheets/United Nations (1994) Earth Summit. Convention on Desertifica-showFS. php?number=2 [Geo-2–171]

tion. United Nations Conference on Environment and Develop-ment, Rio de Janeiro, Brazil, 3–14 June 1992. In. United Nations,Vitousek PM, Mooney HA, Lubchenco J, Melillo JM (1997) HumanNew York, p 44

Vogel CH, Smith J (2002) Building social resilience in arid ecosys-domination of Earth’s ecosystems. Science 277:494–499

tems. In: Reynolds JF, Stafford Smith DM (eds) Global Deserti-fication: Do Humans Cause Deserts? Dahlem University Press,von Hardenberg J, Meron E, Shachak M, Zarmi Y (2001) DiversityBerlin, pp 149–166

of vegetation patterns and desertification. Phys Rev Let 87:Whisenant SG (1999) Repairing Damaged Wildlands. Cambridge198101–1–198101–4

White RPUniversity Press, Cambridge

and services: A web-based geospatial analysis. W, Nackoney J (2003) Drylands, people, and ecosystem goodsorld ResourcesWhite R, Tunstall D, Henninger N (2003) An Ecosystem Ap-Institute, Washington, DC

proach to Drylands: Building Support for New DevelopmentPolicies. World Resources Institute, EarthTrends PublicationSeries. Available online: /features/Whitford WG (1993) Animal feedbacks in desertification: An over-view_feature.cfm?fid=37&theme=39

Whitford WG (1997) Desertification and animal biodiversity in theview. Rev Chil Hist Nat 66:243–251

Zaman S (1997) Effects of rainfall and grazing on vegetation yielddesert grasslands of north America. J Arid Environ 37:709–720and cover of two arid rangelands in Kuwait. Environ Cons 24:Zhao WZ, Xiao HL, Liu ZM, Li J (2005) Soil degradation and resto-344–350

ration as affected by land use change in the semiarid Bashangarea, northern China. Catena 59:173–186

本文来源：https://www.bwwdw.com/article/qtd4.html