Improving The Accuracy of Satellite-Based Near-Surface Air Temperature and Precipitation Products

Date

2023-5-2

Author

Karaman, Çağrı Hasan

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

134
views

81
downloads

Cite This

This study evaluates the performance of several reanalyses and satellite-based products of precipitation and near-surface air temperature in estimating daily and monthly variables across the complex terrain of Turkey. The study used 1120 ground-based gauge stations from 2015 to 2019, covering a range of complex topography with different climate classes and land surface types. Various distance-based interpolation, classical Random Forest (RF), and more innovative Random Forest Spatial Interpolation (RFSI) downscaling algorithms were applied to improve the spatial resolution of the products, and several satellite-based covariates were investigated as proxies for downscaling. The results showed that AgERA5 was the best-performing product for air temperature estimation, while MSWEP V2.2 was superior for precipitation estimation. Spatial downscaling using bicubic interpolation improved air temperature product performance, and the RF algorithm outperformed all other methods in certain seasons. However, downscaling algorithms had a negligible effect on precipitation product performance. Moreover, merging of ground-based measurements and several precipitation products were utilized using RFSI algorithm. Additionally, the performance of the merged product was investigated using lumped hydrological modelling approach. The study suggests that combining ground-based measurements, precipitation products, and features related to topography can substantially improve the representation of spatiotemporal precipitation distribution in data-scarce regions. Overall, the study highlights the importance of using alternative sources of data to traditional in-situ observations to obtain accurate estimates of precipitation and air temperature over vast and unmeasured areas.

Subject Keywords

Machine Learning, Near-Surface Air Temperature, Satellite, Precipitation, Downscaling

URI

https://hdl.handle.net/11511/104182

Collections

Graduate School of Natural and Applied Sciences, Thesis