Synthesis

INTRODUCTION: LAND DEGRADATION IN THE CONTEXT OF DISASTER RISK RESEARCH

Land degradation represents one of the most urgent challenges for humanity, because it leads to an often permanent reduction of the land’s biological productivity, yield reduction, or loss, causing economic loss and hunger (Olsson et al. 2019). It adversely affects the well-being of approximately 3.2 billion people worldwide and results in a loss of about 10% of annual gross domestic product (GDP; IPBES 2018). Furthermore, the Millennium Ecosystem Assessment (MA 2005) revealed that around the year 2000 60% of the examined ecosystem services were being degraded or used unsustainably. Nevertheless, the development of effective disaster risk reduction (DRR) strategies in the context of policy and decision making has mostly focused on sudden-onset hazards that appear as shocks, such as floods or earthquakes, while slow-onset events, such as land degradation and drought, have been rarely considered within the conventional disaster risk framework (Matthewman 2015, Zaidi 2018, Hsu 2019, Staupe-Delgado 2019). This synthesis sets out to analyze the discrepancies in risk concepts used in the academic literature on land degradation risk assessment and explores ways to bridge the gap between land degradation research and DRR frameworks.

Over the last 30 years, definitions of land degradation have changed considerably (see Box 1), and according to the Intergovernmental Panel on Climate Change (IPCC) there is still “no single method by which land degradation can be measured objectively and consistently over large areas because it is such a complex and value-laden concept” (Olsson et al. 2019:348). In consequence, there is a wide variation in estimates of the global costs of land degradation, which range from US$17.58 billion to as high as US$9.4 trillion (Nkonya et al. 2016), and its current global extent and severity are still not well quantified (Olsson et al. 2019).

In this paper, we follow the most recent definition proposed by IPCC (Olsson et al. 2019) that defines land degradation as a negative trend in land condition, caused by human-induced processes, expressed as long-term reduction or loss of biological productivity, ecological integrity or value to humans. The processes of land degradation are diverse, encompassing, among others, soil erosion and salinization of the soil, as well as a persistent loss of natural vegetation cover or loss of biodiversity (IPBES 2018, Olsson et al. 2019). All of these processes act on individual temporal and spatial scales, and the dominating processes vary locally. This makes general assessments of land degradation on large scales difficult. The same holds for drivers of land degradation, mainly climate variability (e.g., the frequency of drought or extreme rainfall events) and human activities such as unsustainable agriculture. Finally, processes and drivers are often intermingled (Olsson et al. 2019). The result is that it is often not clear what should be measured, the current state of the ecosystem (e.g., vegetation coverage), the rate of the processes involved (e.g., erosion rate), or the local intensity of the drivers (e.g., intensity of land use, climatic drivers, etc.).

As opposed to sudden-onset hazards that appear as shocks, such as floods or earthquakes, land degradation processes can continue unnoticed for decades before adverse consequences appear (e.g., loss of ecological integrity, biological productivity, or ecosystem services). Research has shown that land degradation dynamics are often non-linear and may unfold gradually as well as suddenly (Turnbull et al. 2008, Karssenberg et al. 2017). Ecological tipping points, points where the system rapidly reorganizes into an alternative system state (Kéfi et al. 2014, Scheffer et al. 2015), can result in rapid cascading changes in ecosystems, even if external drivers change only gradually with time (Scheffer et al. 2001, Scheffer and Carpenter 2003, Turnbull et al. 2008). These non-linear dynamics make ecological regime shifts very hard to predict (Hastings and Wysham 2010), and the consequences may even become apparent only after a regime shift has already happened (Scheffer 1990). Once the ecosystem has been shifted to a degraded functional state, the process is often very difficult and costly to reverse because of positive stabilizing feedback mechanisms (Tomaselli 1977, Scheffer 1990, Rietkerk and van de Koppel 1997, Mäler 2000, Roques et al. 2001).

In consequence, the high damage potential of land degradation has long been concealed and land degradation has rarely been considered as a natural hazard in the context of disaster risk research. The adverse impacts of degradation can be intermingled with the effects of extreme hydrometeorological extreme events, which have often been perceived as the imminent threat in the past, particularly droughts (in this paper we refer to droughts as “the naturally occurring phenomenon that exists when precipitation has been significantly below normal recorded levels, causing serious hydrological imbalances that adversely affect land resource production systems,” UNCCD 1994:5). One prominent example is the prolonged drought in the Sahel region of sub-Saharan Africa in the 1970s and early 1980s that led to the death of over 200,000 people and widespread devastation of rural agriculture including the loss of millions of livestock (UNCCD 2020), providing the background for holding the first United Nations Conference on Desertification (UNCCD) in 1977 (Dregne 2002). The conference officially introduced the term “desertification” as a synonym for land degradation in drylands (Dregne 2002, UNCCD 2020). At first, drought was identified as the main factor for the human Sahel crises in the 1970s. Yet, it soon became clear that drought was not the underlying factor, but that prior land management was a crucial factor in the extent of the damage to the land (Dregne 2002), i.e., it appeared to be a man-made loss in the system’s ecological resilience that had made the land so vulnerable to droughts. In ecology, resilience is defined as a system’s ability to maintain its structural integrity and persist without being pushed into another stable state under the influence of a disturbance (Holling 1973, 1996, Scheffer et al. 2015). The notion that land degradation plays a crucial role in the land’s vulnerability to drought and that drought is often the mere trigger for human catastrophes, not the root cause, has since been a recurring notion in the literature on drought risk (see, e.g., Plate and Merz 2001, Felgentreff and Glade 2007, IPCC 2012, UNDRR 2019).

The relation between land degradation, drought, and human catastrophes exemplifies the finding that when both natural and human vulnerabilities occur at the same time and place, disasters can result (Alcántara-Ayala 2002). In the Sendai Framework of Disaster Risk Reduction 2015–2030 (UNISDR 2015), the United Nations established international procedures for disaster risk reduction (DRR). DRR aims at preventing new and reducing existing disaster risks and increasing resilience (UNDRR 2022). In order to promote a common understanding of DRR concepts and assist the implementation of DRR efforts of authorities, practitioners, and the public, the United Nations Office for Disaster Risk Reduction (UNDRR, formerly UNISDR) developed a coherent terminology related to DRR (Table 1) that is used as a guiding framework in this study.

Understanding risk as a concomitance of hazard, exposure, and vulnerability emphasizes that risk can be managed by modifying its components, i.e., reducing vulnerability or exposure to the hazard (Cardona 2004) and thus helps to identify effective entry points for DRR. It is based on Kaplan and Garrick (1981:13), who suggested that “risk is probability and consequence.” Disaster risk can be derived spatially from a combination of a hazard and a vulnerability map (Plate and Merz 2001); or, it can be represented in the form of risk curves, where the annual frequency of events with a certain magnitude is plotted over their potential overall damage. The unit of risk is then given as the average economic damage or by human loss indicators (loss of lives, people suffering from adverse health effects, etc.) expected per year (Plate and Merz 2001, Merz et al. 2009).

Natural hazards are natural processes or phenomena “that may cause loss of life, injury or other health impacts, property damage, social and economic disruption or environmental degradation” (UNDRR 2022). In disaster risk research the hazard is usually expressed as the probability of occurrence of an event with a certain magnitude at a given location during a specified period of time. The extent to which hazards result in actual damage depends on both the spatial and temporal exposure of the elements at risk and their vulnerability. Vulnerability represents the predisposition of elements that makes them susceptible to the impact of hazards (Cardona 2004, UNDRR 2022) and is expressed in the form of potential damage affecting people, property, or natural capital (Plate and Merz 2001). It depends on the susceptibility (also called sensitivity) as well as the resilience of the affected elements.

Risk assessment forms an important part of DRR, because it constitutes the basis for decision makers’ evaluation of risks. It is described as “a qualitative or quantitative approach to determine the nature and extent of disaster risk by analyzing potential hazards and evaluating existing conditions of exposure and vulnerability that together could harm people, property, services, livelihoods and the environment on which they depend” (UNDRR 2022).

The objective of this synthesis is to analyze the discrepancies in existing land degradation risk assessment approaches, with regard to the disaster risk concept of the UNDRR as summarized in Table 1. By this we want to identify key elements that hinder a link between land degradation risk assessment and land management. We further investigate how the underlying risk concept affects the way results can be linked to decision making in land management. A systematically chosen sample of the scientific literature was analyzed with regard to the underlying risk concept and its components (hazard, exposure, vulnerability), the perspective on land degradation, its processes and drivers, as well as its temporal component. These themes are covered in separate subsections of the results. The studies addressing land degradation with a risk concept that is in line with UNDRR’s definitions are presented separately at the end of the results section. In the discussion, eight key issues currently hindering a link between land degradation risk assessment and land management are identified based on the preceding literature analysis. Finally, based on identified discrepancies, ideas toward a comprehensive land degradation risk assessment are proposed as a way forward, by incorporating the concepts of ecological resilience and disaster risk research.

METHODS

A systematic literature search was conducted in the scientific database of the Web of Science (http://apps.webofknowledge.com). Originally drafted as part of a PhD thesis (von Keyserlingk 2021; date of original literature search: 25 January 2017), the search was extended for this paper to cover the time period from 1900 to 22 December 2021 and the text revised. A set of title search criteria was chosen to identify studies explicitly aiming at a land degradation risk assessment: “(‘land degradation’ OR desertification) AND risk AND (analys* OR assess*).” The keyword “desertification” was included in addition to “land degradation,” because by definition it refers to land degradation in drylands (UNCCD 1994), and the two terms are often used synonymously. The additional terms “analys*” or “assess*” were included, for if only the term “risk” was used, a manual examination of the abstracts revealed that the vast majority of results were not related to a systematic risk assessment approach. Rather, the term “risk” was used in its colloquial meaning. For the same reason, a title search was chosen over a topic search. Only original research articles available in the English language were included. The systematic search resulted in a database of 33 studies covering 30 years (1992–2021, Appendix 1, Table A1). This selection of studies is not intended to be exhaustive of all research performed on the topic, but should present an objectively chosen sample as a basis for the analysis.

The focus of the analysis was to understand in which context the term “risk” was used in each study and how the underlying risk concept affects the possibilities for potentially linking results of the risk analysis with decision making in land management. Hence, the first step was to assess whether risk was used with an (explicit or implicit) underlying risk concept or whether it was used in its colloquial meaning. In a second step, the risk concept was analyzed with regard to the concept of disaster risk used by the UNDRR (Table 1). Studies were loosely grouped according to their understanding of risk (Table A1) to promote a joint discussion of studies on selected thematic aspects related to the underlying risk concept. In the results section, studies are presented with regard to the following questions: In which context was the term “risk” used? What was the underlying risk concept and terminology including the risk components of hazard and vulnerability? Was there a clear distinction between risk, hazard, and vulnerability, between cause and effect? How was the temporal component of land degradation risk addressed? Which studies understood risk as a concomitance of hazard and vulnerability (i.e., similar to the disaster risk concept used by UNDRR) and how was this implemented in the risk analysis?

RESULTS: IDENTIFICATION OF KEY ISSUES

Ambiguity of the term “risk”

The process of searching for studies that conceptually address risk assessment related to land degradation or desertification first of all revealed that the term “risk” was frequently used with the intention of alerting the reader to the seriousness of the problem of land degradation and to the extent of the associated damage to human livelihoods, without any linkage to a conceptual risk framework. Hence, it proved challenging to develop a search strategy by which studies that at least mostly address risk within a conceptual context could be efficiently extracted. The title search criteria described above proved sufficient to meet this goal. Still, seven of the 33 studies (Rubio and Bochet 1998, Taguas et al. 2013, Vorovencii 2015, 2016, 2017, Abdallah and Massoud 2018, Türkes et al. 2020; see Table A1) do not provide any explanation of their conceptual understanding of risk, even though all of them refer to risk assessment for degradation or desertification in the title. The study by Rubio and Bochet (1998) aims at discussing and selecting suitable indicators for a desertification risk assessment. With the expression “desertification risk assessment,” they refer to an “evaluation system of desertification.” In Taguas et al. (2013), the aim stated in the title, a land degradation risk assessment, does not connect to the rest of the article, which is concerned with the calculation of annual soil erosivity from precipitation records. Although this approach takes into account that degradation is a process rather than a fixed state, it does not aim at estimating future risks. Within the text, the terms “risk” and “soil erosion risk” are used with the intention to highlight the potential negative consequences arising from soil erosion. The two studies by Vorovencii (2015, 2016) explicitly aim at a risk assessment. Again, risk assessment is interpreted as an assessment of desertification status (here, based on satellite-derived vegetation greenness, landscape pattern, and surface albedo), which was repeated for several points in time. Additionally, in the study from 2016, the convergence to degraded land is assessed over time; yet, this is not included in their calculation of risk classes. The term “risk” is used very frequently in both studies within various phrases connected to desertification (e.g., “the risk of exposure to desertification” [Vorovencii 2015:7], “exposed to desertification risk” [Vorovencii 2015:9], “measures for fighting against the land degradation risk” [Vorovencii 2016:15], but also to drivers of desertification (e.g., “the risk of destructuring” or “the risk of overgrazing” [in Vorovencii 2016:11,13]). The different ways in which “risk” is used reveals that it is not related to any conceptual framework. Further, there is no clear distinction between hazard and risk, and the hazard is not clearly identified. For example, the hazard could be the desertification processes or it could be drivers of desertification, such as overgrazing. The damaging consequences of desertification are clearly stressed as translating to, among others, a reduction or loss of the biological potential of the land (Vorovencii 2015). In summary, the studies that do not relate the term “risk” to any conceptual framework use it with a colloquial meaning with the intention of stressing the threatening nature of land degradation. All of them appear to confuse a risk assessment with an assessment of the land degradation status or of certain degradation processes.

Risk concepts and terminology: risk, hazard and vulnerability

Twenty-five of the analyzed studies use the term “risk” within a conceptual, risk-related framework (see Table A1). However, most of them do not provide an explicit definition of risk (exceptions are: Ubugunov et al. 2011, Ali and Abdel Kawy 2013, Hai et al. 2013, Wang et al. 2015, Akbari et al. 2016, Salunkhe et al. 2018, Martínez-Valderrama et al. 2020, Masoudi et al. 2021a, 2021b; see Table A1). Three studies (Ubugunov et al. 2011, Wang et al. 2015, Akbari et al. 2016; see Table A1) can be related to the risk concept of UNDRR used in disaster risk research and are discussed separately at the end of the results section. For the other 22 studies, a conceptual framework could be either explicitly or implicitly discerned (see Table A1 for an overview of the respective risk concepts). The analysis revealed that the inherent risk concepts and terminology were not consistent among and within studies. Four of the studies (Masoudi and Jokar 2018, Abu Seif and El-Khashab 2019, Masoudi et al. 2021a, 2021b; see Table A1) do not distinguish between “hazard” and “risk.” In Masoudi et al. (2021a) and in Masoudi et al. (2021b) risk is defined as the probability of hazard occurrence. In the title of Abu Seif and El-Khashab (2019) a “desertification risk assessment” is stated, yet later the “desertification hazard of sand dunes” (p. 359) is assessed. Even more frequently, “risk” is used synonymously with “vulnerability” (namely in: Salvati et al. 2008, Salvati and Zitti 2009, Ladisa et al. 2012, Becerril-Piña et al. 2015, Salvati et al. 2015, Tombolini et al. 2016, Uzuner and Dengiz 2020). For example, Salvati et al. (2008) use desertification risk synonymously with “environmental vulnerability.” In Salvati et al. (2015) spatial vulnerability patterns are used to assess desertification risk. Furthermore, the term “vulnerability” is recurrently used in substitution with “susceptibility” or “sensitivity” (e.g., in Ladisa et al. 2012). In Vieira et al. (2021) “desertification risk” is used synonymously with “desertification susceptibility.” Garg and Harrison (1992) as well as Ali and Abdel Kawy (2013) use “erosion risk” synonymously with “susceptibility to erosion.”

This ambiguity in terminology indicates that no consistent framework including the different risk components exists in the academic land degradation literature and that they are not clearly distinguished from each other. The confusion of risk with vulnerability is frequently associated with an analysis implemented with a geographic information system (GIS), such as the Environmental Sensitive Areas (ESA) framework (based on the EU-funded MEDALUS [Mediterranean Desertification and Land Use] project) - one of the most common approaches to determine desertification risk (Salvati and Zitti 2009). It has been applied by eight studies (Salvati et al. 2008, Salvati and Zitti 2009, Ladisa et al. 2012, Tombolini et al. 2016, Karamesouti et al. 2018, Salunkhe et al. 2018, Pishyar et al. 2020, Uzuner and Dengiz 2020). In this methodology, an Environmental Sensitive Area Index (ESAI) is calculated based on four thematic indicators representing soil quality, climate, vegetation, and management practices / land use (Ladisa et al. 2012, Tombolini et al. 2016). Sometimes, additional socioeconomic factors such as human pressure or demographic trends are added (namely in: Salvati et al. 2008, Ladisa et al. 2012, Salunkhe et al. 2018). The ESAI is calculated as the geometric mean of these indices. The results are generally presented in spatially explicit risk maps identifying different classes of areas “sensitive to desertification” ranging from “non-affected” to “critical” (e.g., in Ladisa et al. 2012). What appears striking is that the ESAI is explicitly called a “risk index,” e.g., by Salvati et al. (2008), and is commonly used to perform “desertification risk assessment,” while at the same time it is widely recognized as a measure of vulnerability to desertification (Salvati and Zitti 2009, Tombolini et al. 2016). The naming of the index itself and its categories refer to environmentally sensitive areas. This highlights the confusion of the terms risk, vulnerability, and sensitivity.

Different perspectives on land degradation: how to distinguish between processes and drivers, cause and effect, hazard and vulnerability

What remains generally unclear is what the land is vulnerable to. Not only is there usually no distinction between hazard and vulnerability, but also the perspectives on land degradation vary among studies. Whereas some studies focus on soil-related processes (e.g., Garg and Harrison 1992, Ali and Abdel Kawy 2013, Taguas et al. 2013), the majority of studies (e.g., Ibáñez et al. 2008, Becerril-Piña et al. 2015, Wang et al. 2015, Weinzierl et al. 2016) follow the more modern, holistic view on land degradation, including different components of the human-environment system (see an overview of land degradation definitions in Box 1). Based on this perspective, it has become common practice to look at the bulk impact of land degradation processes rather than assessing individual processes (Ibáñez et al. 2008). According to Salvati and Zitti (2009:960), the ESA approach does match this requirement exactly, because it “does not focus on a specific process of land degradation ... but quantifies the synergic impact of different factors potentially leading to land degradation.” Other risk indices follow a similar approach. For example, Santini et al. (2010) combine vegetation, soil properties, soil erosion, seawater intrusion, and grazing models to calculate a Desertification Risk Index. Similarly, Karavitis et al. (2020) calculate a desertification risk index (DRI) based on a combination of indicators related to the physical environment, land management, and socioeconomic factors. Weinzierl et al. (2016) calculate an Integrated Degradation Risk Index based on topography, land cover, soil, land use, demography, infrastructure, and climate. Although approaches such as this one indeed account in detail for many factors associated with land degradation, they create another problem: When all factors are merged within one index, it becomes impossible to differentiate between drivers (e.g., climatic variation and land use) and processes (e.g., soil erosion rate, seawater intrusion, etc.) of land degradation. Additionally, the bio-physical susceptibility (e.g., topography, soil and vegetation properties) of the land itself cannot be disentangled. Last, expected damage cannot be quantified, for it is intermingled within the risk index.

Among the analyzed literature, several studies effectively distinguishing between cause and effect in land degradation risk were identified. Those that can be related to the risk concept used by UNDRR are discussed separately in the last part of the results section. Other approaches were found in Ali and Abdel Kawy (2013), Hai et al. (2013), Ibáñez et al. (2008), and Martínez-Valderrama et al. (2020), and are discussed below. The modeling study by Ibáñez et al. (2008) uses system stability analysis to identify long-term land use conditions leading to desertification. Assuming constant climate and economic scenarios, they focus on the human-resource system. Resource exploitation is founded on economic decisions. Simulations revealed that high-profit scenarios are responsible for the expansion of desertification. The model clearly distinguishes between the drivers (here: human activity) and the result (here: loss in provision of resources). Thereby, Ibáñez et al. (2008) indirectly interpret human overexploitation of resources as the hazard, and the loss in provision of resources, i.e., land degradation, as the potential damage (i.e., the vulnerability). Another interesting, yet very different approach that clearly differentiates between cause and effect was found in Hai et al. (2013). Here, a probabilistic understanding of risk is expressed and risk is defined as “the chance of something happening that will have impacts on objectives” (Hai et al. 2013:1546). The authors used a semi-qualitative approach focusing on cause-effect relationships that are arranged in a risk assessment matrix (“Leopold Matrix” after Leopold et al. 1971). Individual socioeconomic activities that are potentially affected by desertification, such as rice or cotton cultivation, raising livestock, or water supply, were collected together with a set of environmental and socioeconomic factors associated with desertification (such as land use, topography, rainfall, population density, etc.) potentially affecting those activities. Risk scores were calculated for each target activity based on categories representing the likelihood of occurrence of particular causes (i.e., “hazards,” ranging from “very unlikely” to “almost certain”) in combination with ranks of consequences (i.e., “vulnerability,” ranging from “no impact” to “catastrophic”). The risk calculation is hence based on a combination of likelihood and consequence, and as such shows striking parallels to quantitative risk assessment. However, the assignment of the categories for likelihood as well as for consequences is not based on quantitative data. Rather, stakeholder knowledge and risk perceptions were used as the main input to the risk matrix following a qualitative social sciences approach. Risk scores are not used to quantify expected damage, but to determine which environmental and socioeconomic factors pose the highest threat to socio-economic activities. As such, the approach of using a risk matrix can still assist management decision making, even though it lacks objectivity and precision. Finally, the study by Ali and Abdel Kawy (2013) indirectly estimates the degree to which land use causes soil degradation. Soil degradation risk is calculated by multiplying soil, topographic, and climate ratings in GIS, resulting in a risk map. They differentiate between the “degradation risk,” which they also call the “susceptibility to degradation” (Ali and Abdel Kawy 2013:2774), and the “degradation hazard,” here meaning the actual presence of degradation processes (salinization, sodification, water logging). The difference between these two is attributed to human impact (positive or negative); i.e., land that has a low degradation risk, but is actually strongly degraded indicates land mismanagement, while land with a high degradation risk, but a low degradation status is managed beneficially. Thereby, inappropriate land use practices can be identified.

These three examples illustrate that a differentiation between cause (hazard) and effect (vulnerability) directly creates a link to decision making in land management. However, the diversity of approaches and lack of a shared risk concept and terminology hamper coherent land degradation risk assessment.

Temporal component of land degradation risk

Several studies (namely: Salvati and Zitti 2009, Santini et al. 2010, Becerril-Piña et al. 2015, Salvati et al. 2015, Tombolini et al. 2016, Vorovencii 2016, 2017, Weinzierl et al. 2016, Karamesouti et al. 2018, Masoudi et al. 2021a, 2021b) consider temporal trends in their analysis, thereby accounting for the dynamic nature of land degradation, as well as for the potential of future risk. Santini et al. (2010:395) point out in their introduction that “an area mildly degraded subject to an intense and rapid desertification process is more at risk when compared to an area at higher degradation level but stable over time.” They account for this by including temporal degradation trends in their risk calculation. Yet, in the end, the seriously degraded areas are assigned to the “very high risk class,” whereas critical areas at a “precarious equilibrium between the natural environment and human activities” are assigned to the “high risk class” (Santini et al. 2010:406). This classification appears contradictory to their statement in the introduction. An integration of static and temporal components within one risk index (as in Santini et al. 2010 or in Weinzierl et al. 2016) makes it impossible to disentangle future risk from the present degradation status. Another approach to account for the dynamic nature of risk is a spatiotemporal convergence analysis, as performed by Salvati and Zitti (2009), Tombolini et al. (2016), and Vorovencii (2016). Here, spatial convergence of land degradation status or vulnerability is calculated over time to estimate future degradation trends, which is interpreted as risk. This approach indicates that risk is seen as something dynamic, different from a degradation status assessment. Becerril-Piña et al. (2015) base their risk assessment purely on the temporal component of risk: They did not include any static factors such as terrain or soil properties, but analyzed changes in temporal trends of land degradation factors (vegetation, climate, and anthropogenic). Ibáñez et al. (2008) and Martínez-Valderrama et al. (2020) treat risk of desertification as the likelihood that an undesirable system state is reached in the long term, thereby clearly addressing desertification risk as something potentially happening in the future. Martínez-Valderrama et al. (2020) used a state and transition model of social-ecological systems and assessed risk based on the percentage of simulations in which the final stable state was undesirable (key resources exhausted or critical thresholds crossed). Three studies (Masoudi and Jokar 2018, Masoudi et al. 2021b, 2021a) clearly distinguish between the present degradation status (termed “actual risk” in Masoudi and Jokar 2018) and “potential risk” of degradation in the future. Furthermore, they introduce a statistical picture of risk by assigning probability levels to those areas at potential risk of becoming further degraded in the future.

In conclusion, the incorporation of a temporal component in risk estimation is a necessary step that promotes, but does not necessarily lead to, a clear distinction between land degradation status and future risk. Accounting for a probabilistic nature of risk, as has been done in more recent studies, promotes such a distinction, thereby facilitating a link to land management. In contrast, assessing risk based on changes in temporal degradation trends alone disregards other elements of land degradation risk, such as vulnerability, thereby hindering a link to land management. The scenario-based approach applied in state and transition models is a very valuable step toward understanding the effects of land management decisions on land degradation risk over the long term.

Land degradation risk assessment approaches related to the disaster risk concept of UNDRR

The approaches attempting a land degradation risk assessment that can be linked to the disaster risk concept of UNDRR (namely: Ubugunov et al. 2011, Wang et al. 2015, Akbari et al. 2016; see Table A1) have in common that all of them include an explicit risk definition that clearly distinguishes hazard from vulnerability. The hazard is seen as consisting of anthropogenic factors, such as agricultural activities or livestock grazing, as well as natural factors, including climatic effects such as precipitation, evaporation, and wind. Some properties of the land itself, such as vegetation coverage (in Wang et al. 2015) are included here too, but they also form a part of the vulnerability (e.g., soil properties in Wang et al. 2015). In Ubugunov et al. (2011) the properties of the land itself are described as endogenous hazards, whereas climate and human activities are described as exogenous hazards. At the same time, Ubugunov et al. (2011) repeatedly use the phrase “desertification hazard,” indicating that the occurrence of desertification processes themselves is seen as the hazard. In line with this, they consider areas exposed to certain desertification processes (water erosion, deflation, salinization) in their risk analysis (see Table 1 in Ubugunov et al. 2011:182). In Akbari et al. (2016), desertification severity is referred to as representing the hazard. However, the assessment of desertification severity is also referred to as the “calculation of the risk” (Ubugunov et al. 2011:370). This shows that the terms “risk” and “hazard” are not clearly distinguished from each other.

With the vulnerability term, potential damage is considered. In the three studies discussed here, vulnerability refers to a potential loss in ecosystem productivity resulting in economic losses, i.e., by considering the fraction of agricultural land affected by land degradation as in Ubugunov et al. (2011), or by including the ratios of the agricultural population and agricultural outputs to the gross domestic product (GDP) as in Wang et al. (2015). Additionally, the vulnerability term includes ecosystem properties, such as vegetation cover (Akbari et al. 2016), and soil properties (Wang et al. 2015), which make land vulnerable to degradation processes. This shows that vulnerability is approached on two levels: first, on the socioeconomic level, and second, on the ecosystem level. In Wang et al. (2015) restorability is considered separately, yet purely in a socioeconomic context (i.e., referring to the level of education and to a variety of measures for disaster prevention such as rehabilitation measures). In Akbari et al. (2016) it does not become clear what the vulnerability term consists of. Vulnerability is described as a “function of stability (resilience) and instability (vulnerability),” as well as a “function of resilience and sensitivity of the elements at risk” (Akbari et al. 2016:369). The first definition appears to be inconsistent, for it defines vulnerability as a function of vulnerability itself. A vulnerability value is assigned based on social-ecological conditions, but also on the presence of risk itself, for they write that “elements that are at higher risk classes will be more vulnerable” (Akbari et al. 2016:369, see also their Table 4). This appears to be circular reasoning. With “risk class” they seem to refer to the desertification severity, again confusing risk and hazard.

The dynamic component of land degradation risk is approached in contrasting ways. Akbari et al. (2016) do not consider a dynamic risk component at all. In Ubugunov et al. (2011), the inherently probabilistic nature of both hazards and vulnerability is explicitly acknowledged: risk is quantified as expected annual damage. Yet, at the same time, the authors understand desertification as a permanent hazard existing “during the entire length of time of its action” (Ubugunov et al. 2011:180) and keep the probability term constant over time. Thus, although the inclusion of a probabilistic term within a risk model is a novelty of this study, its implementation has obvious shortcomings. In Wang et al. (2015:1712) the hazard term is described as the “variation degree of natural disaster and human factors which caused the desertification disaster.” Even though no probabilistic term is explicitly included, risk is seen as something dynamic, depending on the variation of natural disasters. Yet, this dynamic idea is not coherently followed up on in their analysis (except that some of the input datasets they used are continuous, e.g., meteorological data, normalized difference vegetation index with 16-day resolution). Interestingly, in contrast to Ubugunov et al. (2011), they describe desertification as a “typical disaster risk event” (Wang et al. 2015:1703), which suggests a discrete event in time.

In Ubugunov et al. (2011) the results of their risk analysis are presented quantitatively, giving direct estimates for the different administrative districts of the Republic of Buryatia: in affected hectares/year for the physical desertification risk, in rubles/year for the economic risk, and in the number of affected persons/year. It is the only study where expected damages are quantitatively assessed. Both in Akbari et al. (2016) and in Wang et al. (2015) risk is presented in the form of an index, with risk categories ranging from low to very high risk. Separate maps for the different components of risk (hazard, elements at risk, vulnerability [+ restorability in Wang et al. 2015]), as well as final risk maps are produced. The study by Wang et al. (2015) is unique in clearly presenting which factors were included in which risk component (see Fig. 4 in Wang et al. 2015:1713).

The approaches applying a natural disaster risk assessment to land degradation show that even though they share a similar risk concept, the terminology is inconsistently used within and among studies. Vulnerability and hazard are still difficult to disentangle. Only in Wang et al. (2015) is it clearly discernible what factors the different components of risk are composed of. None of the studies present a comprehensive solution for capturing the dynamic nature of land degradation adequately to allow projections of future risks. This highlights the difficulty of including a probabilistic approach in land degradation risk assessment. Often, the analysis falls short of the concept.

DISCUSSION: WHAT IS THE DILEMMA OF CURRENT LAND DEGRADATION RISK ASSESSMENT APPROACHES?

Land degradation describes a variety of processes that lead to a reduction of the land’s biological productivity, integrity, or value to humans (Olsson et al. 2019). Hence, the definition itself contains an explicit inclusion of ecological and socioeconomic damage resulting from these processes (see Box 1). This provides a direct link to the vulnerability concept as an expression of potential damage resulting from a hazard, as it is used in the context of disaster risk research. Curiously, however, to the present day no consistent framework for a quantitative assessment of land degradation risk appears to have been developed: The synthesis presented here has revealed that while approaches to calculate land degradation risk have evolved to include and capture ever more processes and factors involved in land degradation and risk assessment in detail, the development of a coherent conceptual risk framework is trailing behind. At the same time, an urgent need to derive “measures for fighting against the land degradation risk” (Vorovencii 2016:15) appears to be generally well-recognized. A quantification of that risk, however, is hampered by an overall discrepancy in what this risk contains. The analysis of 33 studies dealing with land degradation risk assessment approaches revealed that not even a consensus exists on what is perceived as the hazard. Altogether, eight key issues could be identified, hampering the development of a stringent and coherent risk assessment for land degradation and a link to risk-informed decision making:

1. The term “risk” is frequently used in its colloquial meaning, with the intention of alerting and emphasizing the seriousness of the problem of land degradation and the extent of potential damage to human livelihoods. This emphasizes that the perception of land degradation as a threat to humanity is there, even though it is not connected to a formal risk concept.
2. Risk assessment is often confused with a degradation or desertification status or process assessment (i.e., monitoring of land degradation). This confusion hinders a link to the different phases of disaster risk management (i.e., prevention, preparedness, response, and recovery; UNISDR (2015)) and a projection of future risks. This is an important barrier to linking risk analysis to land management options and interventions.
3. Risk terminology is not consistently used among and within studies. The terms risk, hazard, vulnerability, and susceptibility are often confused. Only a minority of studies (Ubugunov et al. 2011, Hai et al. 2013, Wang et al. 2015, Akbari et al. 2016, Salunkhe et al. 2018, Martínez-Valderrama et al. 2020) include an explicit risk definition. Even if a risk theory exists, it is not consistent between studies and often the analysis falls short of the concept.
4. Risk assessment and vulnerability assessment are often confused and used synonymously. Vulnerability is usually not linked to potential damage to the people (exceptions are: Ubugunov et al. 2011, Hai et al. 2013, Wang et al. 2015), but rather used as a descriptor for the land’s likelihood of becoming degraded. This likelihood is assessed based on a combination of environmental susceptibility to degradation and the present land degradation status (plus past trends), as well as on climatic and human drivers. The confusion of risk and vulnerability assessment makes it impossible to differentiate between present land degradation status and future risks.
5. Potential damage (i.e., average annual economic or ecological damage, number of injured persons, or loss of lives) is mentioned frequently as a motivation or to raise attention to the urgency of the problem (similar to how “risk” is often used). Yet, damage is rarely explicitly considered or quantified in the risk analyses (exceptions are: Ubugunov et al. 2011, Hai et al. 2013, Wang et al. 2015). Only one study (Ubugunov et al. 2011) quantified risk by estimating the expected annual damage.
6. There is no consensus around the “hazard” land degradation. The definitions of land degradation have evolved in historical contexts. This is mirrored in land degradation risk assessment approaches. What is perceived as the threat (i.e., the hazard) ranges from pure soil-related processes to attempts that include a large variety of processes adversely affecting the human-environment system. The fact that the hazard is usually not specified hinders the predictability of hazard occurrence.
7. In the majority of studies, there is no clear distinction between processes and drivers, cause and effect, hazard and vulnerability (exceptions are: Ibáñez et al. 2008, Ubugunov et al. 2011, Ali and Abdel Kawy 2013, Hai et al. 2013, Wang et al. 2015, Akbari et al. 2016, Martínez-Valderrama 2020). This hampers the development of quantitative risk projections, as well as a link to decision making and land management. Even in studies that explicitly include hazard and vulnerability terms, a clear differentiation between the two remains difficult (exception is: Wang et al. 2015).
8. Temporal dynamics are sometimes considered, but mostly not embedded within a theoretical risk framework. Usually, land degradation is treated as either permanent or a discrete hazard. However, some studies (Ibáñez et al. 2008, Ubugunov et al. 2011, Masoudi and Jokar 2018, Martínez-Valderrama et al. 2020, Masoudi et al. 2021a, 2021b) do include a probabilistic element of risk, facilitating a distinction between present degradation status and future risk. Two of those studies (Ibáñez et al. 2008, Martínez-Valderrama et al. 2020) are based on scenario analyses in state and transition models.

OUTLOOK

In the following, we propose ideas to approach bridging the gap between land degradation research and DRR concepts in order to stimulate debate and further research on this globally relevant topic. We argue that integrating the already internationally accepted risk framework outlined by the UNDRR could help overcome several of the above-identified shortcomings in land degradation risk assessment approaches and substantially strengthen the link to land management and risk-informed decision making.

First of all, using consistent risk terminology of UNDRR (2022) would promote the transferability of results and the implementation of DRR strategies by decision makers. Second, our literature analysis indicated that it is necessary to clearly differentiate between drivers and processes of land degradation. Only then does a differentiation between hazard and vulnerability, between present degradation status and future risk become possible, thereby promoting a better understanding of risk and an identification of effective entry points for DRR measures (Cardona 2004, Hessel et al. 2014, Masoudi et al. 2021b). Stating the respective perspective on land degradation (e.g., which definition is used) helps to establish clarity about what is actually perceived as a risk. In order to connect to the disaster risk framework of UNDRR, where risk is described as a function of exposure, hazard, and vulnerability, we propose to identify the key land degradation processes that reduce the land’s ecological integrity, productivity, or value to humans as perceived by land users on a local or regional level and treat them as the hazards, henceforth “degradation hazards” (Table 2, Fig. 1).

In the next step, the specific drivers of the involved processes as well as the associated ecological and socioeconomic vulnerability (potential damage) should be identified. The adverse effects on the ecosystem (here: “ecological vulnerability”) depend on the bio-physical susceptibility and on the ecological resilience to the degradation hazards. Previous research has demonstrated that human actions (such as land management) have the potential to reduce ecological resilience and that a loss of resilience increases the likelihood of regime shifts toward a less desirable (i.e., degraded) ecosystem state (Holling 1973, Gunderson 2000, Scheffer et al. 2001, Folke et al. 2004, Vetter 2009). Gunderson (2000) argues that if ecosystems shift to a less desirable (i.e., degraded) system state, resource crises occur. Hence, the concept of ecological resilience offers valuable entry points for a quantification of ecological vulnerability to degradation processes and climate variability under different land use options. We propose to build on the concept of ecological resilience to clearly distinguish between rapid climatic drivers (e.g., extreme rainfall events or droughts) of land degradation hazards and gradual drivers of degradation (e.g., land use) that affect the ecological resilience to these degradation hazards and thereby act on the ecological vulnerability (Fig. 2).

Treating the dynamic components of land degradation drivers in two ways, i.e., probabilistically for rapid climatic drivers, and scenario-based for slow anthropogenic drivers and long-term climatic trends, may present a way out of the dilemma of treating land degradation either as a permanent hazard, which yields few options for a probabilistic approach as exemplified in Ubugunov et al. (2011), or a discrete, but unpredictable event in time. Incorporating the steps suggested in Figure 1 into land degradation risk assessment approaches would promote a quantification of the area-specific land degradation risk resulting from concrete degradation hazards in combination with the underlying ecological vulnerability. For a given probability of hazard occurrence, several land management and climate scenarios could then be compared (as shown for flood risk in Thieken et al. 2016). This is an important step toward empirically based projections of land degradation risk and a link to risk-informed decision making in land management. Because the need for a coupling of natural and human vulnerabilities has been stressed in several studies (e.g., Alcántara-Ayala 2002, Turner et al. 2003), a valuable next step would be the inclusion of socioeconomic damage.

CONCLUSIONS

In this paper we aimed at critically analyzing existing land degradation risk assessment approaches and identifying discrepancies with regard to the internationally established risk concept of UNDRR. Key identified shortcomings are that the term “risk” is frequently used in its colloquial meaning, and that risk terminology is not consistent across and within studies. A main challenge hindering a link to land management is that there is often no distinction between a degradation status assessment or a degradation monitoring approach and an assessment of future risk. This difficulty is further amplified by a confusion of risk assessment with an assessment of the land’s vulnerability to degradation. Damage is rarely explicitly considered or quantified, and there appears to be no consensus about what the hazard of land degradation contains. In the majority of studies there is no clear distinction between processes and drivers, cause and effect, hazard and vulnerability. Finally, temporal dynamics of land degradation are rarely considered in a manner that promotes a link to land management. However, in more recent studies there appears to be a development toward accounting for the probabilistic nature of degradation risk. This development promotes a distinction between an assessment of present degradation status and future risk. Altogether, this synthesis identified several key limitations of existing land degradation risk assessment approaches, thereby offering valuable entry points for establishing a link to risk-informed land management. Finally, we argue that integrating the already internationally established risk framework outlined by the UNDRR could help overcome several of the above-identified shortcomings. To advance this idea, conceptual ideas toward bridging the gap between land degradation research and DRR strategies were developed. First, we propose to follow consistent risk terminology outlined by UNDRR (UNDRR 2022) when addressing land degradation risk. Second, we emphasize the need to clearly differentiate between hazard, vulnerability, and risk, as well as between drivers and processes of land degradation. By differentiating the various components of risk, existing and very detailed risk assessment frameworks for land degradation such as the frequently applied Environmental Sensitive Areas (ESA) approach could be linked more directly to different stages of disaster risk management (i.e., prevention, preparedness, response, and recovery), as suggested in the Sendai Framework of Disaster Risk Reduction (UNISDR 2015). Further, we propose to clearly distinguish between rapid climatic drivers of land degradation hazards and gradual drivers such as land use affecting the land’s ecological resilience to these hazards and thereby the ecological vulnerability. Hence, effects of different climate and land management scenarios could be compared to assess future land degradation risk. These ideas can be valuable starting points to facilitate further discussions on this globally relevant topic and for the development of effective risk reduction measures for land degradation.

LITERATURE CITED

Abdallah, S., and E. Massoud. 2018. Land degradation risk assessment in Al-Sawda terraces, Kingdom of Saudi Arabia. Arabian Journal of Geosciences 11:599. https://doi.org/10.1007/s12517-018-3956-x

Abu Seif, E., and M. El-Khashab. 2019. Desertification risk assessment of sand dunes in Middle Egypt: a geotechnical environmental study. Arabian Journal for Science and Engineering 44(1):357-375. https://doi.org/10.1007/s13369-018-3343-7

Akbari, M., M. Ownegh, H. R. Asgari, A. Sadoddin, and H. Khosravi. 2016. Desertification risk assessment and management program. Global Journal of Environmental Science and Management 2(4):365-380. https://doi.org/10.22034/gjesm.2016.02.04.006

Alcántara-Ayala, I. 2002. Geomorphology, natural hazards, vulnerability and prevention of natural disasters in developing countries. Geomorphology 47(2-4):107-124. https://doi.org/10.1016/S0169-555X(02)00083-1

Ali, R. R., and W. A. M. Abdel Kawy. 2013. Land degradation risk assessment of El Fayoum depression, Egypt. Arabian Journal of Geosciences 6(8):2767-2776. https://doi.org/10.1007/s12517-012-0524-7

Becerril-Piña, R., C. A. Mastachi-Loza, E. González-Sosa, C. Díaz-Delgado, and K. M. Bâ. 2015. Assessing desertification risk in the semi-arid highlands of central Mexico. Journal of Arid Environments 120:4-13. https://doi.org/10.1016/j.jaridenv.2015.04.006

Cardona, O. D. 2004. The need for rethinking the concepts of vulnerability and risk from a holistic perspective: a necessary review and criticism for effective risk management. Chapter 3 in G. Bankoff, D. Hilhorst, and G. Frerks, editors. Mapping vulnerability: disasters, development and people. Earthscan, Routledge, London, UK.

Dregne, H. E. 2002. Land degradation in the drylands. Arid Land Research and Management 16(2):99-132. https://doi.org/10.1080/153249802317304422

Felgentreff, C., and T. Glade. 2007. Naturrisiken und Sozialkatastrophen. Spektrum Akademischer Verlag, Heidelberg, Germany

Folke, C., S. Carpenter, B. Walker, M. Scheffer, T. Elmqvist, L. Gunderson, and C. S. Holling. 2004. Regime shifts, resilience, and biodiversity in ecosystem management. Annual Review of Ecology, Evolution, and Systematics 35(1):557-581. https://doi.org/10.1146/annurev.ecolsys.35.021103.105711

Garg, P. K., and A. R. Harrison. 1992. Land degradation and erosion risk analysis in S.E. Spain: a geographic information system approach. Catena 19(5):411-425. https://doi.org/10.1016/0341-8162(92)90041-9

Gunderson, L. H. 2000. Ecological resilience - in theory and application. Annual Review of Ecology and Systematics 31(1):425-439. https://doi.org/10.1146/annurev.ecolsys.31.1.425

Hai, L. T., A. Gobin, and L. Hens. 2013. Risk assessment of desertification for Binh Thuan Province, Vietnam. Human and Ecological Risk Assessment: An International Journal 19(6):1544-1556. https://doi.org/10.1080/10807039.2012.716688

Hastings, A., and D. B. Wysham. 2010. Regime shifts in ecological systems can occur with no warning. Ecology Letters 13(4):464-472. https://doi.org/10.1111/j.1461-0248.2010.01439.x

Hessel, R., M. S. Reed, N. Geeson, C. J. Ritsema, G. van Lynden, C. A. Karavitis, G. Schwilch, V. Jetten, P. Burger, M. J. van der Werff ten Bosch, S. Verzandvoort, E. van den Elsen, and K. Witsenburg. 2014. From framework to action: the DESIRE approach to combat desertification. Environmental Management 54(5):935-950. https://doi.org/10.1007/s00267-014-0346-3

Holling, C. S. 1973. Resilience and stability of ecological systems. Annual Review of Ecology and Systematics 4(1):1-23. https://doi.org/10.1146/annurev.es.04.110173.000245

Holling, C. S. 1996. Engineering resilience versus ecological resilience. Pages 31-43 in P. C. Schulze, editor. Engineering within ecological constraints. National Academies, Washington, D.C., USA.

Hsu, E. L. 2019. Must disasters be rapidly occurring? The case for an expanded temporal typology of disasters. Time & Society 28(3):904-921. https://doi.org/10.1177/0961463X17701956

Ibáñez, J., J. Martínez-Valderrama, and J. Puigdefábregas. 2008. Assessing desertification risk using system stability condition analysis. Ecological Modelling 213(2):180-190. https://doi.org/10.1016/j.ecolmodel.2007.11.017

Intergovernmental Panel on Climate Change (IPCC). 2012. Managing the risks of extreme events and disasters to advance climate change adaptation: special report of the Intergovernmental Panel on Climate Change. C. B. Field, V. Barros, T. F. Stocker, and Q. Dahe, editors. Cambridge University Press, Cambridge, UK. https://doi.org/10.1017/CBO9781139177245

Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services (IPBES). 2018. The IPBES assessment report on land degradation and restoration. L. Montanarella, R. Scholes, and A. Brainich, editors. Secretariat of the IPBES, Bonn, Germany. https://doi.org/10.5281/zenodo.3237392

Kaplan, S., and B. J. Garrick. 1981. On the quantitative definition of risk. Risk Analysis 1(1):11-27. https://doi.org/10.1111/j.1539-6924.1981.tb01350.x

Karamesouti, M., P. Panagos, and C. Kosmas. 2018. Model-based spatio-temporal analysis of land desertification risk in Greece. Catena 167:266-275. https://doi.org/10.1016/j.catena.2018.04.042

Karavitis, C., D. Tsesmelis, P. Oikonomou, O. Kairis, C. Kosmas, V. Fassouli, C. Ritsema, R. Hessel, V. Jetten, N. Moustakas, B. Todorovic, N. Skondras, C. Vasilakou, S. Alexandris, E. Kolokytha, D. Stamatakos, R. Stricevic, E. Chatzigeorgiadis, J. Brandt, N. Geeson, and G. Quaranta. 2020. A desertification risk assessment decision support tool (DRAST). Catena 187:104413. https://doi.org/10.1016/j.catena.2019.104413

Karssenberg, D., M. F. P. Bierkens, and M. Rietkerk. 2017. Catastrophic shifts in semiarid vegetation-soil systems may unfold rapidly or slowly. American Naturalist 190(6). https://doi.org/10.1086/694413

Kéfi, S., V. Guttal, W. A. Brock, S. R. Carpenter, A. M. Ellison, V. N. Livina, D. A. Seekell, M. Scheffer, E. H. van Nes, and V. Dakos. 2014. Early warning signals of ecological transitions: methods for spatial patterns. PLoS ONE 9(3):e92097. https://doi.org/10.1371/journal.pone.0092097

Land Degradation Assessment in Drylands (LADA). 2011. Land degradation assessment in drylands: methodology and results. Food and Agriculture Organization of the United Nations, Rome, Italy. https://www.fao.org/3/i3241e/i3241e.pdf

Ladisa, G., M. Todorovic, and G. Trisorio Liuzzi. 2012. A GIS-based approach for desertification risk assessment in Apulia region, SE Italy. Physics and Chemistry of the Earth, Parts A/B/C 49:103-113. https://doi.org/10.1016/j.pce.2011.05.007

Leopold, L. B., F. E. Clarke, B. B. Hanshaw, and J. R. Balsley. 1971. A procedure for evaluating environmental impact. U.S. Geological Survey, Washington, D.C., USA. https://doi.org/10.3133/cir645

Mäler, K.-G. 2000. Development, ecological resources and their management: a study of complex dynamic systems. European Economic Review 44:645-665. https://doi.org/10.1016/S0014-2921(00)00043-X

Martínez-Valderrama, J., J. Ibáñez, F. Alcalá, and S. Martínez. 2020. SAT: a software for assessing the risk of desertification in Spain. Scientific Programming 2020:7563928. https://doi.org/10.1155/2020/7563928

Masoudi, M., M. Elhaeesahar, and A. Cerdà . 2021a. Risk assessment of land degradation (RALDE) in Khuzestan Province, Iran. Eurasian Soil Science 54(8):1228-1240. https://doi.org/10.1134/S1064229321080135

Masoudi, M., and P. Jokar. 2018. Risk assessment of desertification using GIS in upper and lower reaches of Mond basin, Iran. AUC Geographica 53(2):156-166. https://doi.org/10.14712/23361980.2018.16

Masoudi, M., M. Vahedi, and A. Cerda. 2021b. Risk assessment of land degradation (RALDE) model. Land Degradation & Development 32(9):2861-2874. https://doi.org/10.1002/ldr.3883

Matthewman, S. 2015. Disasters, risks and revelation: making sense of our times. Palgrave Macmillan, London, UK. https://doi.org/10.1057/9781137294265

Merz, B., F. Elmer, and A. H. Thieken. 2009. Significance of "high probability/low damage" versus "low probability/high damage" flood events. Natural Hazards and Earth System Sciences 9(3):1033-1046. https://doi.org/10.5194/nhess-9-1033-2009

Millennium Ecosystem Assessment (MA). 2005. Ecosystems and human well-being: synthesis. Island Press, Washington, D.C., USA.

Nachtergaele, F., R. Biancalani, S. Bunning, and H. George. 2010. Land degradation assessment: the LADA approach. Pages 72-75 in 19th World Congress of Soil Science, Symposium 3.1.1, Further development of soil evaluation methods, Soil Solutions for a Changing World, 1–6 August, Brisbane, Australia.

Nkonya, E., A. Mirzabaev, and J. von Braun, editors. 2016. Economics of land degradation and improvement - a global assessment for sustainable development. Springer, Cham, Switzerland. https://doi.org/10.1007/978-3-319-19168-3

Olsson, L., H. Barbosa, S. Bhadwal, A. Cowie, K. Delusca, D. Flores-Renteria, K. Hermans, E. Jobbagy, W. Kurz, D. Li, D. J. Sonwa, L. Stringer. 2019. Land degradation. Chapter 4 in P. R. Shukla, J. Skea, E. Calvo Buendia, V. Masson-Delmotte, H.-O. Pörtner, D. C. Roberts, P. Zhai, R. Slade, S. Connors, R. van Diemen, M. Ferrat, E. Haughey, S. Luz, S. Neogi, M. Pathak, J. Petzold, J. Portugal Pereira, P. Vyas, E. Huntley, K. Kissick, M. Belkacemi, J. Malley, editors. Climate change and land: an IPCC special report on climate change, desertification, land degradation, sustainable land management, food security, and greenhouse gas fluxes in terrestrial ecosystems. https://www.ipcc.ch/srccl/chapter/chapter-4/

Pishyar, S., H. Khosravi, A. Tavili, A. Malekian, and S. Sabourirad. 2020. A combined AHP- and TOPSIS-based approach in the assessment of desertification disaster risk. Environmental Modeling & Assessment 25(2):219-229. https://doi.org/10.1007/s10666-019-09676-8

Plate, E. J., and B. Merz, editors. 2001. Naturkatastrophen: Ursachen, Auswirkungen, Vorsorge. E. Schweizerbart’sche Verlagsbuchhandlung (Nägele u. Obermüller), Stuttgart, Germany.

Popkin, G. 2014. On the edge. Science 345(6204):1552-1554. https://doi.org/10.1126/science.345.6204.1552

Rietkerk, M., and J. van de Koppel. 1997. Alternate stable states and threshold effects in semi-arid grazing systems. Oikos 79(1):69-76. https://doi.org/10.2307/3546091

Roques, K. G., T. G. O'Connor, and A. R. Watkinson. 2001. Dynamics of shrub encroachment in an African savanna: relative influences of fire, herbivory, rainfall and density dependence. Journal of Applied Ecology 38(2):268-280. https://doi.org/10.1046/j.1365-2664.2001.00567.x

Rubio, J. L., and E. Bochet. 1998. Desertification indicators as diagnosis criteria for desertification risk assessment in Europe. Journal of Arid Environments 39(2):113-120. https://doi.org/10.1006/jare.1998.0402

Salunkhe, S., A. Bera, S. Rao, V. Venkataraman, U. Raj, and Y. Murthy. 2018. Evaluation of indicators for desertification risk assessment in part of Thar Desert Region of Rajasthan using geospatial techniques. Journal of Earth System Science 127:116. https://doi.org/10.1007/s12040-018-1016-2

Salvati, L., C. Kosmas, O. Kairis, C. Karavitis, S. Acikalin, A. Belgacem, A. Solé-Benet, M. Chaker, V. Fassouli, C. Gokceoglu, H. Gungor, R. Hessel, H. Khatteli, A. Kounalaki, A. Laouina, F. Ocakoglu, M. Ouessar, C. Ritsema, M. Sghaier, H. Sonmez, H. Taamallah, L. Tezcan, and J. de Vente. 2015. Unveiling soil degradation and desertification risk in the Mediterranean basin: a data mining analysis of the relationships between biophysical and socioeconomic factors in agro-forest landscapes. Journal of Environmental Planning and Management 58(10):1789-1803. https://doi.org/10.1080/09640568.2014.958609

Salvati, L., and M. Zitti. 2009. Convergence or divergence in desertification risk? Scale-based assessment and policy implications in a Mediterranean country. Journal of Environmental Planning and Management 52(7):957-971. https://doi.org/10.1080/09640560903181220

Salvati, L., M. Zitti, and T. Ceccarelli. 2008. Integrating economic and environmental indicators in the assessment of desertification risk: a case study. Applied Ecology and Environmental Research 6(1):129-138. https://doi.org/10.15666/aeer/0601_129138

Santini, M., G. Caccamo, A. Laurenti, S. Noce, and R. Valentini. 2010. A multi-component GIS framework for desertification risk assessment by an integrated index. Applied Geography 30(3):394-415. https://doi.org/10.1016/j.apgeog.2009.11.003

Scheffer, M. 1990. Multiplicity of stable states in freshwater systems. Hydrobiologia 200:475-486. https://doi.org/10.1007/BF02530365

Scheffer, M., S. Carpenter, J. A. Foley, C. Folke, and B. Walker. 2001. Catastrophic shifts in ecosystems. Nature 413(6856):591-596. https://doi.org/10.1038/35098000

Scheffer, M., and S. R. Carpenter. 2003. Catastrophic regime shifts in ecosystems: linking theory to observation. Trends in Ecology & Evolution 18(12):648-656. https://doi.org/10.1016/j.tree.2003.09.002

Scheffer, M., S. R. Carpenter, V. Dakos, and E. H. van Nes. 2015. Generic indicators of ecological resilience: Inferring the chance of a critical transition. Annual Review of Ecology, Evolution, and Systematics 46(1):145-167. https://doi.org/10.1146/annurev-ecolsys-112414-054242

Staupe-Delgado, R. 2019. Progress, traditions and future directions in research on disasters involving slow-onset hazards. Disaster Prevention and Management: An International Journal 28(5):623-635. https://doi.org/10.1108/DPM-11-2018-0358

Taguas, E. V., E. Carpintero, and J. L. Ayuso. 2013. Assessing land degradation risk through the long-term analysis of erosivity: a case study in southern Spain. Land Degradation & Development 24(2):179-187. https://doi.org/10.1002/ldr.1119

Thieken, A. H., H. Cammerer, C. Dobler, J. Lammel, and F. Schöberl. 2016. Estimating changes in flood risks and benefits of non-structural adaptation strategies - a case study from Tyrol, Austria. Mitigation and Adaptation Strategies for Global Change 21(3):343-376. https://doi.org/10.1007/s11027-014-9602-3

Tomaselli, R. 1977. The degradation of the Mediterranean maquis. Ambio 6(6):356-362. https://www.jstor.org/stable/4312322

Tombolini, I., A. Colantoni, G. Renzi, A. Sateriano, A. Sabbi, N. Morrow, and L. Salvati. 2016. Lost in convergence, found in vulnerability: a spatially-dynamic model for desertification risk assessment in Mediterranean agro-forest districts. Science of The Total Environment 569-570:973-981. https://doi.org/10.1016/j.scitotenv.2016.06.049

Türkes, M., T. Öztaş, E. Tercan, G. Erpul, A. Karagöz, O. Dengiz, O. Doğan, K. Şahin, and B. Avcioğlu. 2020. Desertification vulnerability and risk assessment for Turkey via an analytical hierarchy process model. Land Degradation & Development 31(2):205-214. https://doi.org/10.1002/ldr.3441

Turnbull, L., J. Wainwright, and R. E. Brazier. 2008. A conceptual framework for understanding semi-arid land degradation: ecohydrological interactions across multiple-space and time scales. Ecohydrology 1(1):23-34. https://doi.org/10.1002/eco.4

Turner II, B. L., R. E. Kasperson, P. A. Matson, J. J. McCarthy, R. W. Corell, L. Christensen, N. Eckley, J. X. Kasperson, A. Luers, M. L. Martello, et al. 2003. A framework for vulnerability analysis in sustainability science. Proceedings of the National Academy of Sciences 100(14):8074-8079. https://doi.org/10.1073/pnas.1231335100

Ubugunov, L. L., A. I. Kulikov, and M. A. Kulikov. 2011. On the application of risk analysis technology for assessment of the ecological hazard of desertification (by the example of Republic of Buryatia). Contemporary Problems of Ecology 4(2):178-185. https://doi.org/10.1134/S1995425511020093

United Nations Convention to Combat Desertification (UNCCD). 1994. United Nations Convention to Combat Desertification (UNCCD) in those countries experiencing serious drought and/or desertification, particularly in Africa. United Nations General Assembly, Paris, France. https://www.unccd.int/sites/default/files/2022-02/English_0.pdf

United Nations Convention to Combat Desertification (UNCCD). 2020. United Nations Convention to Combat Desertification. UNCCD, Bonn, Germany. https://www.unccd.int

United Nations International Strategy for Disaster Risk Reduction (UNISDR). 2015. Sendai framework for disaster risk reduction 2015-2030. The Framework was adopted at the Third UN World Conference on Disaster Risk Reduction in Sendai, Japan, on March 18, 2015. https://www.undrr.org/publication/sendai-framework-disaster-risk-reduction-2015-2030

United Nations Office for Disaster Risk Reduction (UNDRR). 2019. Global assessment report on disaster risk reduction. UNDRR, Geneva, Switzerland.

United Nations Office for Disaster Risk Reduction (UNDRR). 2022. Terminology on disaster risk reduction. UNDRR, Geneva, Switzerland. https://www.undrr.org/terminology

Uzuner, C., and O. Dengiz. 2020. Desertification risk assessment in Turkey based on environmentally sensitive areas. Ecological Indicators 114:106295. https://doi.org/10.1016/j.ecolind.2020.106295

Vetter, S. 2009. Drought, change and resilience in South Africa's arid and semi-arid rangelands. South African Journal of Science 105:29. https://doi.org/10.4102/sajs.v105i1/2.35

Vieira, R., J. Tomasella, A. Barbosa, M. Martins, D. Rodriguez, F. Rezende, F. Carriello, and M. Santana. 2021. Desertification risk assessment in Northeast Brazil: current trends and future scenarios. Land Degradation & Development 32(1):224-240. https://doi.org/10.1002/ldr.3681

von Keyserlingk, J. 2021. Ecological resilience and the risk of land degradation in a dry Mediterranean rangeland. Dissertation. Technische Universität Berlin, Berlin, Germany. https://doi.org/10.14279/depositonce-12496

von Keyserlingk, J., M. de Hoop, A. G. Mayor, S. C. Dekker, M. Rietkerk, and S. Foerster. 2021. Resilience of vegetation to drought: studying the effect of grazing in a Mediterranean rangeland using satellite time series. Remote Sensing of Environment 255:112270. https://doi.org/10.1016/j.rse.2020.112270

Vorovencii, I. 2015. Assessing and monitoring the risk of desertification in Dobrogea, Romania, using Landsat data and decision tree classifier. Environmental Monitoring and Assessment 187:204. https://doi.org/10.1007/s10661-015-4428-3

Vorovencii, I. 2016. Assessing and monitoring the risk of land degradation in Baragan Plain, Romania, using spectral mixture analysis and Landsat imagery. Environmental Monitoring and Assessment 188:439. https://doi.org/10.1007/s10661-016-5446-5

Vorovencii, I. 2017. Applying the change vector analysis technique to assess the desertification risk in the south-west of Romania in the period 1984–2011. Environmental Monitoring and Assessment 189:524. https://doi.org/10.1007/s10661-017-6234-6

Wang, Y., J. Zhang, E. Guo, and Z. Sun. 2015. Fuzzy comprehensive evaluation-based disaster risk assessment of desertification in Horqin Sand Land, China. International Journal of Environmental Research and Public Health 12(2):1703-1725. https://doi.org/10.3390/ijerph120201703

Weinzierl, T., J. Wehberg, J. Böhner, and O. Conrad. 2016. Spatial assessment of land degradation risk for the Okavango River catchment, Southern Africa. Land Degradation & Development 27(2):281-294. https://doi.org/10.1002/ldr.2426

Zaidi, R. Z. 2018. Beyond the Sendai indicators: application of a cascading risk lens for the improvement of loss data indicators for slow-onset hazards and small-scale disasters. International Journal of Disaster Risk Reduction 30:306-314. https://doi.org/10.1016/j.ijdrr.2018.03.022