New advanced analytic methods applied to soil data for enhancing Earth System Modeling.
A deep neural network method can learn soil moisture responses to atmospheric forcings and the patterns and conditions of land surface model errors, and by utilizing them, can extend soil moisture data decades back without auxiliary observations.
The application of this new reflection method to Earth system codes has the potential to greatly enhance model productivity and code-reusability.
Soil moisture is a crucial variable affecting runoff, evapotranspiration, energy balance, and ecosystem functioning. It is simulated by most land surface models. Soil moisture observations can be used to evaluate model simulations and reduce uncertainties. The Soil Moisture Active Passive (SMAP) mission has delivered valuable sensing of surface soil moisture since 2015. However, it has a short time span with irregular revisit schedules. Deep learning techniques recently gained rapid adoption and are promoting transformational changes across many disciplines. However, it has not been widely employed in hydrology or Geosciences. Utilizing a state-of-the-art time series deep learning neural network, Long Short-Term Memory (LSTM), DOE-funded researchers created a system that predicts SMAP level-3 moisture product with atmospheric forcings, model-simulated moisture, and static physiographic attributes as inputs. The system removes most of the bias with model simulations and improves predicted moisture climatology, achieving small test root-mean-square errors (<0.035) and high-correlation coefficients >0.87 for over 75% of Continental United States, including the forested southeast. The study showed the proposed network avoids overfitting and is robust for both temporal and spatial extrapolation tests. It also showed that the inclusion of numerical models in a big data machine learning setting helps reduce prediction bias. LSTM generalizes well across regions with distinct climates and environmental settings. With high fidelity to SMAP, LSTM shows great potential for hindcasting, model evaluation, data assimilation, and weather forecasting. This study is the first use of state-of-the-art time series deep learning for hydrology or hydrologic variables.
Contacts (BER PMs)
Earth System Modeling
Penn State University
Partial support from the U.S. Department of Energy Office of Science, Biological and Environmental Research, Earth System Modeling Program.
Fang, K., C. Shen, D. Kifer, and X. Yang. “Prolongation of SMAP to Spatio-temporally Seamless Coverage of Continental US Using a Deep Learning Neural Network.” Geophysical Research Letters (2017). [DOI: 10.1002/2017GL075619]
SC-23.1 Climate and Environmental Sciences Division, BER
BER supports basic research and scientific user facilities to advance DOE missions in energy and environment. More about BER
May 10, 2019
Quantifying Decision Uncertainty in Water Management via a Coupled Agent-Based Model
Considering risk perception can improve the representation of human decision-making processes in age [more...]
May 09, 2019
Projecting Global Urban Area Growth Through 2100 Based on Historical Time Series Data and Future Scenarios
Study provides country-specific urban area growth models and the first dataset on country-level urba [more...]
May 05, 2019
Calibrating Building Energy Demand Models to Refine Long-Term Energy Planning
A new, flexible calibration approach improved model accuracy in capturing year-to-year changes in bu [more...]
May 03, 2019
Calibration and Uncertainty Analysis of Demeter for Better Downscaling of Global Land Use and Land Cover Projections
Researchers improved the Demeter model’s performance by calibrating key parameters and establi [more...]
Apr 22, 2019
Representation of U.S. Warm Temperature Extremes in Global Climate Model Ensembles
Representation of warm temperature events varies considerably among global climate models, which has [more...]
List all highlights (possible long download time)