Biomass Productivity-Based Mapping of Global Land Degradation Hotspots

Abstract

CitationLe Abstract Land degradation affects negatively the livelihoods and food security of global population. There have been recurring efforts by the international community to identify the global extent and severity of land degradation. Using the long-term trend of biomass productivity as a proxy of land degradation at global scale, we identify the degradation hotspots in the world across major land cover types. We correct factors confounding the relationship between the remotely sensed vegetation index and land-based biomass productivity, including the effects of inter-annual rainfall variation, atmospheric fertilization and intensive use of chemical fertilizers. Our findings show that land degradation hotpots cover about 29 % of global land area and are happening in all agro-ecologies and land cover types. This figure does not include all areas of degraded lands, it refers to areas where land degradation is most acute and requires priority actions in both in-depth research and management measures to combat land degradation. About 3.2 billion people reside in these degrading areas. However, the number of people affected by land degradation is likely to be higher as more people depend on the continuous flow of ecosystem goods and services from these affected areas. Land improvement has occurred in about 2.7 % of global land area during the last three decades, suggesting that with appropriate actions land degradation trend could be reversed. We also identify concrete aspects in which these results should be interpreted with cautions, the limitations of this work and the key areas for future research. (Oldeman et al. 1990;USDA-NRCS 1998;Eswaran et al. 2001). The earlier generation of these studies had been constrained by lack of global level quantitative data which could be used for mapping soil and land degradation, and therefore were based on expert opinions. The developments in the remote sensing and satellite technologies allowed the later studies to be based on quantitative satellite data, such as Global Inventory Modelling and Mapping Studies (GIMMS) dataset of 64 km 2 -resolution of Normalized Difference Vegetation Index (NDVI) data, however, several methodological challenges still exist on more accurately estimating the land degradation hotspots (Vlek et al. 2010;Le et al. 2012).In this context, addressing land degradation may require channeling substantial amounts of scarce resources and making long-term investments. These investments are likely to yield high levels of social returns and welfare improvements. However, all countries in the world have budgetary constraints, necessitating the prioritization of such investments. To combat land degradation, both on the international and national levels, policy makers often need information about areas of severe degradation in order to prioritize national budgets and plan strategic interventions (Vlek et al. 2010;Vogt et al. 2011;Le et al. 2012). To achieve this, accurate maps of land degradation hotspots-where land degradation is most acute, are needed. T...

Economics of sustainable land management

“…To illustrate, [42,44] conduct global level analyses of the drivers of land degradation using broadly the same set of explanatory variables, considering the same time period between 1982 and 2006, and in both cases their dependent variables are obtained from the Normalized Difference Vegetation Index (NDVI) from the Global Inventory Modeling and Mapping Studies (GIMMS) [46]. However, [42] use the values of the NDVI directly as given in the NDVI database, while [44] use these values after processing by [10 ], who remove from the NDVI values potential biases emanating from rainfall dynamics, atmospheric and chemical fertilization 3 . The results obtained by the two studies, consequently, have significant divergences due to the substantial differences between their dependent variables in identification of degraded areas.…”

“…12 Environmental change issues 3 Le et al [10] do this by, first, identifying statistically significant trend in NDVI time series between 1982 and 2006, followed by masking of the pixels where NDVI changes are correlated with and are likely to be driven by rainfall dynamics rather than anthropogenic unsustainable management. After removal of the effect of rainfall dynamics, Le et al [10] also mask the pixels indicating pristine areas with insignificant human intervention, where NDVI increases are likely to have occurred due to atmospheric fertilization [47].…”

“…After removal of the effect of rainfall dynamics, Le et al [10] also mask the pixels indicating pristine areas with insignificant human intervention, where NDVI increases are likely to have occurred due to atmospheric fertilization [47]. Finally, the areas with chemical fertilizer application (e.g.…”

“…For instance, some well-known points of debate on the drivers of land degradation include whether higher population causes land degradation [33][34][35], or leads to SLM [36]; whether poverty is the driver of land degradation [37][38][39], or not [40]; would higher market access lead to SLM [41,42], or to land degradation [43]. Mirzabaev et al [44], using newly available datasets at global level on land degradation hotspots [10 ], and various socioeconomic and institutional indicators, find that the causal patterns of individual drivers of land degradation vary across geographic locations. Similar conclusions are also reached by [3 ,4].…”

“…Land degradation has occurred on about 30% of global land area since early 1980s [10 ], while land improvement has occurred only on about 3% of the global land area during the same period [10 ]. The drivers of land degradation are numerous and often have context-specific characteristics.…”