Forecasting Seasonal Streamflow Using a Stacked Recurrent Neural Network
Abstract
Providing accurate seasonal (0-6 months) forecasts of streamflow is critical for applications ranging from optimizing water management to hydropower generation. In this study we evaluate the performance of stacked Long Short Term Memory (LSTM) neural networks, which maintain an internal set of states and are therefore well-suited to modeling dynamic processes.
Existing LSTM models applied to hydrological modeling use all available historical information to forecast contemporaneous output. This modeling approach breaks down for long-term forecasts because some of the observations used as input are not available in the future (e.g., from remote sensing). To solve this deficiency we trained a stacked LSTM model where the first network accumulates the internal states and cells using historical data. These states and cells are then used to initialize the second LSTM which uses meteorological forcings to create streamflow forecasts at various horizons. This method allows the model to learn general hydrological relationships in the temporal domain across different catchment types and project them into the future up to 6 months ahead. Using meteorological time series from NOAA's Climate Forecast System (CFS), remote sensing data including snow cover, vegetation and surface temperature from NASA's MODIS sensors, static catchment attributes, and streamflow data from USGS we trained a stacked LSTM model on hundreds of basins, and evaluated predictions on out-of-sample periods from these same basins. We performed sensitivity analysis on the effects of remote sensing data and static catchment attributes to understand the informational content of these various inputs under various model architectures. Finally, we benchmarked our model to forecasts derived from climatological averages.- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2020
- Bibcode:
- 2020AGUFMH188...02L
- Keywords:
-
- 1816 Estimation and forecasting;
- HYDROLOGY;
- 1817 Extreme events;
- HYDROLOGY;
- 1942 Machine learning;
- INFORMATICS;
- 4337 Remote sensing and disasters;
- NATURAL HAZARDS