Drought monitoring in Yunnan Province based on a TRMM precipitation product

Yunnan Province is a region with frequent droughts; thus, drought monitoring research is important for implementing active and effective measures to mitigate drought and scientifically guide agricultural production. In this study, the Tropical Rainfall Measuring Mission (TRMM 3B43) remote sensing-ba...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Natural hazards (Dordrecht) 2020-12, Vol.104 (3), p.2369-2387
Hauptverfasser: Yu, Yuanhe, Wang, Jinliang, Cheng, Feng, Deng, Huan, Chen, Sheng
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Yunnan Province is a region with frequent droughts; thus, drought monitoring research is important for implementing active and effective measures to mitigate drought and scientifically guide agricultural production. In this study, the Tropical Rainfall Measuring Mission (TRMM 3B43) remote sensing-based product is used as the data source, and a geographically weighted regression (GWR) model, normalized difference vegetation index (NDVI) data and gross primary productivity (GPP) are used as independent variables. The TRMM 3B43 data are downscaled to 1 km spatial resolution to obtain two downscaled precipitation models (GWR_NDVI and GWR_GPP). The precipitation anomaly percentage (Pa) index and the tropical rainfall condition index (TRCI) are used to evaluate the drought situation in Yunnan Province from 2009 to 2018, and the standardized precipitation index (SPI) and standardized precipitation evapotranspiration index (SPEI) are used to verify the Pa and TRCI. The results show the following: (1) With an R 2 as high as 0.8821 and BIAS close to zero, the TRMM 3B43 monthly precipitation is significantly correlated with the measured precipitation. The GWR_NDVI data increase the R 2 at the monthly scale by 0.0114; the GWR_NDVI data show greater improvements from spring and winter than from summer and autumn; and the R 2 of the GWR_NDVI data for some sites are slightly reduced. The R 2 of GWR_GPP data is smaller than that of the TRMM data and GWR_NDVI data at all timescales. (2) Drought occurs every month from 2009 to 2018; it decreases from November to February of the following year and is generally alleviated from March to April; and the incidence of drought from 2009 to 2014 is generally higher than that from 2015 to 2018. The Pa and TRCI show strong correlations with the SPI and SPEI and thus can be used to effectively monitor drought events in Yunnan, although the degree of drought assessed by the Pa and TRCI differs. (3) The spatial distribution of precipitation in Yunnan Province shows little precipitation in the north and east but abundant precipitation in the south and west. Precipitation is mainly concentrated from May to October, with the most abundant precipitation occurring in July.
ISSN:0921-030X
1573-0840
DOI:10.1007/s11069-020-04276-2