Differentiating anthropogenic modification from climate-driven change in Kazakhstan
Ranjeet John, University of South Dakota, ranjeet.john@usd.edu (Presenter)
Venkatesh Kolluru, University of South Dakota, venkatesh.kolluru@coyotes.usd.edu
Jiquan Chen, Michigan State University, jqchen@msu.edu
Anthropogenic activities exert unprecedented stress on dryland ecosystems that encompass 41% of the land surface. Dryland Asia’s grasslands are particularly prone to experiencing large-scale changes in their functioning. However, little is known about how these semiarid ecosystems respond to changes induced by complex human–environmental interactions. Hence, it is important to distinguish the impacts of anthropogenic activities and climate dynamics on vegetation productivity. We used time series of 250m resolution satellite data to show trends in land use and climate variability over the past two decades in Kazakhstan. We implemented TSS-RESTREND and found that 56% of Kazakhstan experienced significant land degradation evident in southern and western provinces. We found that land use change is the predominant contributor to land degradation (26.4%), followed by climate variability (24.7%), climate change (4%) and CO2 fertilization (0.9%). Livestock grazing is a significant driver of land use contributing to land degradation in dryland regions. While traditional livestock numbers are available at the country/provincial level, significant knowledge and data gaps regarding their distribution and impact on degradation exist at the county level. Hence, we aimed to spatially disaggregate district-level livestock numbers into gridded estimates of livestock density using machine learning algorithms. Consequently, a new database of large-scale high-resolution (1 km×1 km) gridded livestock density maps were developed for Kazakhstan from 2000-2019 using vegetation proxies, climatic, socioeconomic, topographic and proximity drivers to explain spatiotemporal trends in grassland degradation. We employed a pixel-wise fitted random forest model and fixed effects model to account for space-time effects and investigate region-specific key socio-environmental system drivers (land use, climatic, topographic and proximity drivers) causing land degradation in Kazakhstan. Our work shows that wall-to-wall estimates of grassland degradation and their drivers can be obtained from exploring space–time data, adding empirical insights to the relationship between land use change and grassland ecosystems in Kazakhstan.
Poster: Poster_John_1-1_110_35.pdf
Associated Project(s):
Poster Location ID: 1-1
Presentation Type: Poster
Session: Poster Session 1
Session Date: Tue (May 9) 5:00-7:00 PM
CCE Program: LCLUC