A fine-resolution soil moisture dataset for China in 2002–2018

Soil moisture is an important parameter required for agricultural drought monitoring and climate change models. Passive microwave remote sensing technology has become an important means to quickly obtain soil moisture across large areas, but the coarse spatial resolution of microwave data imposes great limitations on the application of these data. We provide a unique soil moisture dataset (0.05∘, monthly) for China from 2002 to 2018 based on reconstruction model-based downscaling techniques using soil moisture data from different passive microwave products – including AMSR-E and AMSR2 (Advanced Microwave Scanning Radiometer for Earth Observing System) JAXA (Japan Aerospace Exploration Agency) Level 3 products and SMOS-IC (Soil Moisture and Ocean Salinity designed by the Institut National de la Recherche Agronomique, INRA, and Centre d’Etudes Spatiales de la BIOsphère, CESBIO) products – calibrated with a consistent model in combination with ground observation data. This new fine-resolution soil moisture dataset with a high spatial resolution overcomes the multisource data time matching problem between optical and microwave data sources and eliminates the difference between the different sensor observation errors. The validation analysis indicates that the accuracy of the new dataset is satisfactory (bias: −0.057, −0.063 and −0.027 m3 m−3; unbiased root mean square error (ubRMSE): 0.056, 0.036 and 0.048; correlation coefficient (R): 0.84, 0.85 and 0.89 on monthly, seasonal and annual scales, respectively). The new dataset was used to analyze the spatiotemporal patterns of soil water content across China from 2002 to 2018. In the past 17 years, China's soil moisture has shown cyclical fluctuations and a slight downward trend and can be summarized as wet in the south and dry in the north, with increases in the west and decreases in the east. The reconstructed dataset can be widely used to significantly improve hydrologic and drought monitoring and can serve as an important input for ecological and other geophysical models. The data are published in Zenodo at https://doi.org/10.5281/zenodo.4738556 (Meng et al., 2021a).

A Neural-Network Based Spatial Resolution Downscaling Method for Soil Moisture: Case Study of Qinghai Province

Currently, soil-moisture data extracted from microwave data suffer from poor spatial resolution. To overcome this problem, this study proposes a method to downscale the soil moisture spatial resolution. The proposed method establishes a statistical relationship between low-spatial-resolution input data and soil-moisture data from a land-surface model based on a neural network (NN). This statistical relationship is then applied to high-spatial-resolution input data to obtain high-spatial-resolution soil-moisture data. The input data include passive microwave data (SMAP, AMSR2), active microwave data (ASCAT), MODIS data, and terrain data. The target soil moisture data were collected from CLDAS dataset. The results show that the addition of data such as the land-surface temperature (LST), the normalized difference vegetation index (NDVI), the normalized shortwave-infrared difference bare soil moisture indices (NSDSI), the digital elevation model (DEM), and calculated slope data (SLOPE) to active and passive microwave data improves the retrieval accuracy of the model. Taking the CLDAS soil moisture data as a benchmark, the spatial correlation increases from 0.597 to 0.669, the temporal correlation increases from 0.401 to 0.475, the root mean square error decreases from 0.051 to 0.046, and the mean absolute error decreases from 0.041 to 0.036. Triple collocation was applied in the form of [NN, FY3C, GEOS-5] based on the extracted retrieved soil-moisture data to obtain the error variance and correlation coefficient between each product and the actual soil-moisture data. Therefore, we conclude that NN data, which have the lowest error variance (0.00003) and the highest correlation coefficient (0.811), are the most applicable to Qinghai Province. The high-spatial-resolution data obtained from the NN, CLDAS data, SMAP data, and AMSR2 data were correlated with the ground-station data respectively, and the result of better NN data quality was obtained. This analysis demonstrates that the NN-based method is a promising approach for obtaining high-spatial-resolution soil-moisture data.

Sea–ice concentration at 1 km resolution in summer from merged visible and microwave radiometer observations

<p>Sea–ice concentration, the surface fraction of ice in a given area, is a key component of the Arctic climate system, governing for example the ocean–atmosphere heat exchange. Satellite–based remote sensing offers the possibility for large–scale monitoring of the sea–ice concentration. Using passive microwave measurements, it is possible to observe the sea–ice concentration all year long, almost independently of cloud coverage. The spatial resolution of these measurements is limited to 5 km and coarser. Data from the visible and thermal infrared spectrum offer finer resolutions of 250 m–1 km, but need clear–sky scenes and, in case of visible data, sunlight. In previous work, we developed and analysed a merged dataset of passive microwave and thermal infrared data, combining AMSR2 and MODIS satellite data at 1 km spatial resolution. It has benefits over passive microwave data in terms of the finer spatial resolution and an enhanced potential for lead detection. At the same time, it outperforms thermal infrared data due to its spatially continuous coverage and the statistical consistency with the extensively evaluated passive microwave data. Due to higher surface temperatures in summer, the thermal–infrared based retrieval is limited to winter and spring months. In this contribution, we present first results of extending the existing dataset to summer by using visible data instead of thermal infrared data. The reflectance contrast between ice and water is used for the sea–ice concentration retrieval and results of merging visible and microwave data at 1 km spatial resolution are presented. Difficulties for both, the microwave and visual, data are surface melt processes during summer, which make sea–ice concentration retrieval more challenging. The merged microwave, infrared and visual dataset opens the possibility for a year–long, spatially continuous sea ice concentration dataset at a spatial resolution of 1 km.</p>