LDN assessment at national, subnational and local level
First studies of this cycle were devoted to the results from assessing individual territories of the Russian Federation using the basic Trend.Earth Quantum-GIS module (TE, 2018) or author’s analogs using three global LDN indicators: dynamics of land cover, productivity and soil organic carbon stock (SOC). The results were quite versatile and contradictory. (Trifonova et al., 2021, 2020) focus on the importance of the watershed approach for identifying the boundaries of territories subject to LDN assessment. The main argument is that the greatest homogeneity of biophysical and geochemical conditions is conserved within river basins, which in turn determines the similarity of ecosystems where the effectiveness of LDN mitigation hierarchy of avoid-reduce-reverse (Orr et al. 2017) can be considered:. However, despite the fact that this approach is valid from an environmental perspective, (Andreeva and Kust 2020), (Kust et al. 2019, Kust et al. 2018b) - demonstrated that for national and subnational reporting purposes, it is more efficient to use administrative units, due to a system of relevant statistical data collected by national agencies.
A preliminary assessment for pilot regions of the European part of Russia (Kust, Andreeva, and Lobkovskiy 2020) demonstrated that the LDN concept and the Trend.Earth platform can be considered as a basis for monitoring degradation processes at sub-national level, thus facilitating effective SLM decisions. At the same time, almost all such studies noted that specific data obtained from global databases differ to various extents from ground and statistical data. Also, the Trend.Earth platform, which uses a relatively low data resolution (250-300 m per pixel) resulted problematic for local assessment due to the possible mismatch of boundaries and erroneous interpretation of satellite imagery (Figure 2). Additional studies displayed that the use of Landsat satellite imagery and neural networks for image interpretation, as well as the clarification of land types list, provide an opportunity to increase the mapping accuracy by at least 70 times and simultaneously correct the interpretation of positive and negative dynamics for a specific site (Kust et al. 2021). Although this approach is relatively laborious, it becomes useful for local LDN estimates (Figures 3, 4) (RSF 2020).
National level comparative calculations of the total area of degraded land (SDG indicator 15.3.1) using national statistical data (by summing an area of ‘deteriorated’ land calculated using sectoral methods for different types of land (forest, agricultural, industrial land, etc.)) and Trend.Earth, demonstrated that results obtained by the first method (6.1% of the total land area) are significantly less than obtained through Trend.Earth (12.3%) (Andreeva and Kust 2020). This clearly indicates a need for complex harmonizing between the national and global systems since they use different platforms and calculation methods, indicators and systems for obtaining and monitoring data.
Nevertheless, the first experience of comparative analysis for all 85 Russian federal subjects (administrative units) using Trends.Earth (Andreeva and Kust 2020) demonstrates the high efficiency of this method particularly for comparative studies and the identification of ”hot spots”. For a country like Russia with a high diversity of biophysical and climatic conditions and a variety of economic activities, conducting an overall country assessment of proportion of degraded lands (PDL), with Trend.Earth as a baseline for further calculations (12.3%) does not make sense, since for regions it varies from 67% to less than 1%. In this respect considering the trade-off activities within the whole country as a compensatory scheme for LD does not correspond the LDN principle. Eight federal subjects, where the PDL is less than 2% and the proportion of stable and improved land is high, can be considered as having achieved LDN. At the same time, in another 8 federal subjects with a high concentration of agricultural lands the PDL is very high. From this perspective, a country assessment covering large territories in general may be quite satisfactory and tends to zero in the long term. However, one of the key requirements for LDN is not met: “counter balancing should occur only within individual land types, distinguished by land potential, to ensure “like-for-like” exchanges” (Orr et al. 2017).
(Kust, Andreeva, and Lobkovskiy 2020) proposed using an “LDN Index” for the overall comparative assessment of regions. This index indicates the difference between improved and degraded lands as an intermediate stage in achieving LDN and demonstrates the effectiveness of land policy and practice within a certain territory. (Andreeva and Kust 2020) calculated the Index for all administrative units of the Russian Federation: for Russia in total the LDN Index averages 25.7%, and for individual regions it varies from –57% to +96% (Figure 5). According to this Index, the “hottest” spots in Russia are the territories of the South and Volga federal districts.