A 3 km spatially and temporally consistent European daily soil moisture reanalysis from 2000 to 2015

Abstract

High-resolution soil moisture (SM) information is essential to many regional applications in hydrological and climate sciences. Many global estimates of surface SM are provided by satellite sensors, but at coarse spatial resolutions (lower than 25 km), which are not suitable for regional hydrologic and agriculture applications. Here we present a 16 years (2000-2015) high-resolution spatially and temporally consistent surface soil moisture reanalysis (ESSMRA) dataset (3 km, daily) over Europe from a land surface data assimilation system. Coarse-resolution satellite derived soil moisture data were assimilated into the community land model (CLM3.5) using an ensemble Kalman filter scheme, producing a 3 km daily soil moisture reanalysis dataset. Validation against 112 in-situ soil moisture observations over Europe shows that ESSMRA captures the daily, inter-annual, intra-seasonal patterns well with RMSE varying from 0.04 to 0.06 m 3 m −3 and correlation values above 0.5 over 70% of stations. The dataset presented here provides long-term daily surface soil moisture at a high spatiotemporal resolution and will be beneficial for many hydrological applications over regional and continental scales.

Gap-Free Global Annual Soil Moisture: 15 km Grids for 1991–2018

“…Discrepancy between the augmented ISMN and satellite-derived soil moisture or our downscaled datasets can be associated with differences in the spatial representativeness of points measurements and grids surfaces (Gruber et al, 2020). This scale mismatch has been previously identified when testing different soil moisture patterns (Nicolai-Shaw et al, 2015) as field soil moisture records are usually representative of <1 m 3 of soil while satellite and modeling estimates varies from several meters to multiple kilometers. Soil moisture measurements (from satellites and in situ measurements) across both water-limited environment and tropical areas are extremely limited (Liu et al, 2019), a condition that increases prediction variances (and consequently https://doi.org/10.5194/essd-2020-264 also increased model uncertainty).…”

“…Future work includes predicting global soil moisture patterns across finer pixel sizes (e.g., 1km or <1km) and higher temporal resolutions (e.g., monthly, daily), as it has been done at the regional to continental scales (Naz et al, 2020;Llamas et al, 2020;. The current version of the downscaled soil moisture predictions is provided on an annual basis because is a temporal resolution useful for multiple ecological and hydrological studies related to large-scale ecological processes and climate change (Green et al, 2019).…”

“…The topographic wetness index is a terrain parameter that characterizes areas where soil moisture increase by the effect of overland flow accumulation (Moore et al, 1993). Overland flow and potential incoming solar radiation are two important topographic drivers of the spatial distribution of soil moisture (Nicolai-Shaw et al, 2015), its lags after precipitation events (McColl et al, 2017), and its role as a dominant control of plant productivity (Forkel et al, 2015). Bioclimatic features and soil types account for hydroclimatic and soil variability affecting soil moisture.…”

“…Previous studies have used a variety of prediction factors for soil moisture, including vegetation indexes (from optical imagery), climate information (Alemohammad et al, 2018), chloropeth maps (i.e., land use and land-forms), thermal data and soil information to improve the spatial resolution and coverage of soil moisture gridded datasets (Naz et al, 2020, Peng et al, 2017. In contrast to past efforts, our solution uses a comprehensive set of factors for predicting satellite soil moisture data and independently test the model with in situ soil moisture data.…”

Agricultural Drought Trends on the Iberian Peninsula: An Analysis Using Modeled and Reanalysis Soil Moisture Products

“…However, long-term soil moisture databases are currently available, even globally, that allow for use in applications that were not possible until recently. These databases come from remote sensing [32], modeling or reanalysis [33,34], have been validated with very satisfactory results, and provide decades-long series of soil profile moisture data at everfiner temporal and spatial scales. In the case of remote sensing databases, the series still have a limited length since the launch of satellites dedicated specifically to soil moisture observations, such as Advanced Scatterometer (ASCAT, 2007), Soil Moisture and Ocean Salinity (SMOS, 2010) or Soil Moisture Active Pasive (SMAP, 2015), has been relatively recent.…”