Baseline and status of desertification in Central Asia

Abstract

Baseline determination is a prerequisite for desertification monitoring and assessment and also for the sustainable development of drylands. In this study based on long-term remote sensing net primary production (NPP) data and combined with a local NPP scaling method, the desertification status in Central Asia is quantified, and NPP-based objective baseline of desertification (potential NPP, NPP p ), which could be more convenient and enables the desertification status measurable from local to global scales, is established by integrating space-for-time substitution, isolation forest (iForest) algorithm, and Monte Carlo simulations. More reliable than a fixed percentile-based reference with subjective preference, the analyzed results show that across Central Asia over the past four decades, the NPP p ranged from 46.4 to 639.5 gC m À2 yr À1 with different land cover types and regions, and the general desertification status in the region was rated as a moderate degree with an average of 50.5%. From the 1980s to the 2000s in most regions of Central Asia, reversals of desertification occurred, but it intensified in the 2010s. This study may contribute to improving the existing framework of SDG Indicator 15.3.1.

Analysis of desertification combating needs based on potential vegetation NDVI—A case in the Hotan Oasis

“…Moreover, an 853.65 km 2 area in the study region also indicated a trend of desertification reversal. Our results are in line with Cai et al. (2022) , who reported a significant reversal of desertification within the Hotan oasis, as well as they also proposed PNPP as a benchmark for desertification research in Central Asia as we have used PNDVI as a benchmark for desertification assessment.…”

“…The comparison of PNDVI and ANDVI found that the distribution area of PNDVI with vegetation coverage greater than 15% in the study area increased by 2192.66 km 2 . Our findings are supported by Cai et al (2022), who reported that the natural vegetation in the desert oasis transitional zone in Hotan still has high growth potential. However, contrary to our findings, Pan and Xu (2020) reported that the vegetation coverage in Hotan is less than 60%.…”

“…Desertification has been defined as the land degradation in arid, semi-arid, dry sub-humid areas (collectively known as dryland), resulting from various factors, including climate change and human activity ( The United Nations Convention to Combat Desertification (UNCCD), 1994 ). Desertification has become a global challenge, causing unsustainable land management, and threatening the livelihoods of vulnerable populations ( Barbier and Hochard, 2018 ; Cai et al., 2022 ). Anthropogenic climate change has driven over 5 million km2 of drylands toward desertification over the past three decades, affecting about 213 million people, 93% of them live in developing economies ( Burrell et al., 2020 ).…”

Beyond “greening” and “browning”: Trends in grassland ground cover fractions across Eurasia that account for spatial and temporal autocorrelation

“…Overall, we expected to find only limited coincidence between our soil‐specific results and generic “browning” maps, because most of the previous research considered “browning” as an overall decrease in vegetation indices, not an increase in soil presence, which is relatively sparse in Eurasia (Hu et al, 2020). As such, the detected pixel‐based changes in , and fractions corresponded best with desertification in Central Asia and Kazakhstan arising from agricultural exploitation and climate change (Cai et al, 2022; Hu et al, 2020), and salinization (Robinson, 2016; UNEP, 2011). The high inter‐annual autocorrelation in the detected increase in soil fraction indicates the steadiness of desertification processes, presumably propelled by rising temperatures and aridity (Li et al, 2015).…”

“…Prior studies suggested widespread grassland degradation and desertification in Eurasia (Cai et al, 2022; Hu et al, 2020; Zhang et al, 2018). However, our hypothesis tests found no significant map‐scale decrease in vegetation combined with an increase in the soil fraction (Table 3) even though our trend maps showed pixel‐level decrease in vegetation cover in favor of soil in Central Asia ( and ), and the Gobi Desert () (Figures 3, 4c, and 5b).…”