Bias Correction for ERA5-Land Soil Moisture Product Using Variational Mode Decomposition in the Permafrost Region of Qinghai–Tibet Plateau

Abstract

Soil moisture (SM) is one of the key measures to understand the land-atmosphere interaction and permafrost dynamics in the Qinghai-Tibet Plateau (QTP). ERA5-Land is a new reanalysis product with high spatial resolution (9 km), which can provide long-term SM data with a large spatial coverage as well as at multi-layer soil depths. However, preliminary comparisons with in-situ data show that the ERA5-Land SM product generally underestimates the seasonal variability and demonstrate a positive bias on the QTP. In this paper, we proposed to utilize the mode decomposition method to correct such bias. Specifically, through using the variational mode decomposition (VMD), we decomposed the long time-series of ERA5-Land SM data into a series of Intrinsic Mode Functions (IMFs), and found that the SM seasonal variation can be well represented by the low-frequency modes, which were then selected to feed a regression model for the bias correction. The single-site bias correction results show that our method significantly improves the accuracy of ERA5-Land SM product with bias reduced by 0.22 m 3 /m 3 , 0.31 m 3 /m 3 , 0.15 m 3 /m 3 for alpine meadow, alpine steppe, and alpine desert sites respectively. Together with the slightly reduced accuracy but still acceptable results for the cross-site bias correction, we successfully demonstrate the potential of the mode decomposition method for the bias correction of the ERA5-Land SM product at regional scale. Our method is of great use to study climate impact on regional ecohydrological processes and the permafrost changes in the QTP region.

A New Method of Diagnosing the Historical and Projected Changes in Permafrost on the Tibetan Plateau

“…Then the mean annual data sets are obtained by calculating the arithmetic mean of the data over a year. It is worth noting that the soil moisture content derived from the ERA5‐Land reanalysis data sets does not distinguish the liquid water and solid water content (Chang et al., 2022; Liu & Yang, 2022).…”

“…Then the mean annual data sets are obtained by calculating the arithmetic mean of the data over a year. It is worth noting that the soil moisture content derived from the ERA5-Land reanalysis data sets does not distinguish the liquid water and solid water content (Chang et al, 2022;Liu & Yang, 2022). NDVI (Normalized Difference Vegetation Index) is a dimensionless index to reflect vegetation growth status, which is derived from the global 8 km GIMMS smooth NDVI data set from 1981 to 2015 (Dong & Yang, 2021;Yang et al, 2019).…”