Li, Yifan
(2025).
Modelling biological soil crust based on multiple datasets.
PhD thesis, Universität zu Köln.
This is the latest version of this item.
All available versions of this item
- Modelling biological soil crust based on multiple datasets. (deposited 16 May 2025 10:00) [Currently Displayed]
![]() |
PDF
Thesis_Yifan_Li_published_version.pdf Download (25MB) |
Abstract
Further understanding of biological soil crust (BSC) response to climate change requires BSC-climate models, which represent the relevant processes taking place in the atmosphere and land surface. In this study, a modelling system for biological soil crust and climate factors based on multi-datasets is developed in two approaches. The effects of climate variability on the long temporal and large spatial distribution of BSC are revealed by an improved BSC detection method and multiple linear regression. The models can be used to explain the dominant climatic factors associated with BSC changes. the short-term or long-term forecasts of regional-scale distribution of BSC, the assessment for the potential effects of climate change on the availability of BSC and the sustainable development of ecosystem, as well as the short-term or long-term forecasts of regional-scale distribution of BSC. The long-time and large-scale distribution of biological soil crust is obtained in the study area. To this end, this study is divided into the following four steps: 1) Fusion of MODIS and Landsat7 satellite data using the Sspatial and Temporal Adaptive Reflectance Fusion Model (STARFM) to obtain multispectral data with high spatial and temporal resolution; 2) Calculation of the BSC Index (BSCI) and the NDVI from the fused satellite spectral data; 3) Extraction of the BSC for the study area based on the BSCI thresholds obtained from previous studies, as well as considered with NDVI. 4) Analyzing the extracted BSC data from multiple perspectives. The analysis for 19a shows that on the time scale, the BSC variations have an interannual periodicity, peaking in March and October of each year, and almost zero in winter. On the spatial scale, the BSC is mainly distributed in the desert-oasis transition zone, while the distribution become gradually sparse toward to the desert hinterland. Lag-correlation and partial correlation between BSC and climate variables is analyses. In this study, five climatic variables (specific humidity, 10-meter wind speed, 2-meter temperature, surface solar radiation and total precipitation) and their time lags were used as independent variables. The results show that in some areas the BSC is more strongly correlated with time-lagged climate factors when the time lag is taken into account, and this is most evident for specific humidity. The response of the BSC to this is usually delayed by 1 to 2 months. In principle, the time lag between the BSC and the climatic variables does not exceed three months. The BSC responds quickly to temperature, with a correlation coefficient of 0.7. The BSC also responds quickly to precipitation, while the correlation coefficient is relatively low at 0.46 and the significantly correlated areas are mainly in the east and south. These correlation analyses provide a good reference for the selection of variables for subsequent modelling. The models of biological soil crust and climate factors is constructed using two approaches, in which the influence of time lag is considered. One approach is based on fixed climate factors, and the other slides over the time series to select more appropriate climate factors and coefficients for different time points. Multiple regression analysis is applied to both models. Statistical parameters are used to estimation. The results shows that the two approaches can explain about 40% and 75% of the BSC, respectively. Then applied models to paleoclimate (Last Glacial Maximum and Mid-Holocene) in the Gurbantunggut Desert and to historical climate in the Atacama Desert. Changes in biological soil crust during different time periods are also compared and analyzed. In summary, the long-temporal and large-spatial distribution of BSC is obtained. Benefiting from it, the correlation between BSC and climatic factors is analyzed. And the model system developed captures well the climatological processes in the study area. The BSC-climate model can appropriately predict the BSC in paleoclimate and indicate the its response to the climate variables.
Item Type: | Thesis (PhD thesis) | ||||||||||
Translated abstract: |
|
||||||||||
Creators: |
|
||||||||||
URN: | urn:nbn:de:hbz:38-782917 | ||||||||||
Date: | 2025 | ||||||||||
Language: | English | ||||||||||
Faculty: | Faculty of Mathematics and Natural Sciences | ||||||||||
Divisions: | Faculty of Mathematics and Natural Sciences > Department of Geosciences > Institute for Geophysics and Meteorology | ||||||||||
Subjects: | Natural sciences and mathematics Earth sciences |
||||||||||
Uncontrolled Keywords: |
|
||||||||||
Date of oral exam: | 27 March 2025 | ||||||||||
Referee: |
|
||||||||||
Refereed: | Yes | ||||||||||
URI: | http://kups.ub.uni-koeln.de/id/eprint/78291 |
Downloads
Downloads per month over past year
Export
Actions (login required)
![]() |
View Item |