Latest Carbon Balance Surveys and Monitoring at the Kursk Biosphere Station of the Institute of Geography, Ras, in the Framework of the Novel Concept of Land Degradation Neutrality

Abstract

The paper focuses on the applicability and adaptation of the novel international concept of Land Degradation Neutrality (LDN) of the United Nations, as applied to the territory of Russian Federation, taking into account the large variety of its soil and climatic conditions and economic approaches, as well as the historical past. In particular, in this publication, a team of authors made an attempt to assess to what extent the estimates of the carbon balance and its main components for the study area can be used as one of the indicators of the LDN in relation to the most fertile chernozem agrolandscapes in the south of the European part of Russia (Kursk region). Field work and modelling have been carried out in natural ecosystems and agrolandscapes in the neighbourhood of the Kursk Biosphere Station (KBS) of the Institute of Geography, Russian Academy of Sciences. This valuable russian scientific research station has a long and rich history of geo(eco)system observations with LDN and net C-balance monitoring of age-old agrolandscapes, becoming one of its present focuses. Using simulation models DNDC, RothC, T&P, C-calculators, as well as long-term field measurements of soil CO2 efflux and carbon net exchange, it was found that individual software tools are not able to adequately assess carbon exchange processes of the complex landscape, but correctly selected ensembles consisting of C-balance simulations and C-calculators supported by field data are able to meet such challenges more successfully. On this condition the area's carbon net balance can serve as an alternative to the recommended global LDN indicators.

Development of a methodology for calculating carbon units of heterogeneous territories based on machine learning

“…To date, there are a large number of methods for assessing the carbon balance of the territory [3][4][5]. Most of them are based on the compilation of mathematical models approximating data from point field studies, and partly using machine learning methods.…”