Streamflow Prediction with Limited Spatially-Distributed Input Data

Workshop paper, investigating first ways of using LSTM-based models for climate change related questions in hydrology.

Abstract

Climate change causes more frequent and extreme weather phenomena across the globe. Accurate streamflow prediction allows for proactive and mitigative action in some of these events. As a first step towards models that predict streamflow in watersheds for which we lack ground truth measurements, we explore models that work on spatially-distributed input data. In such a scenario, input variables are more difficult to acquire, and thus models have access to limited training data. We present a case study focusing on Lake Erie, where we find that tree-based models can yield more accurate predictions than both neural and physically-based models.

Paper

Gauch, M. and Mai, J. and Gharari, S. and Lin, J.: “Streamflow Prediction with Limited Spatially-Distributed Input Data”. Workshop on Tackling Climate Change with Machine Learning, 33rd Conference on Neural Information Processing Systems (NeurIPS 2019), Vancouver, Canada.

Workshop Proceedings: Link

Citation

@inproceedings{gauch2019limited,
  title={Streamflow Prediction with Limited Spatially-Distributed Input Data},
  author={Gauch, M. and Mai, J. and Gharari, S. and Lin, J. },
  booktitle={Workshop on Tackling Climate Change with Machine Learning, 33rd Conference on Neural Information Processing Systems (NeurIPS 2019)},
  venue={Vancouver, Canada},
  date={8--14 Dec},
  year={2019}
}