This study proposes a new deep recurrent neural network approach, Neural Runoff Model (NRM), which has been applied on 125 USGS streamflow gages in the State of Iowa for predicting the next 120 h due to the difficult nature of accurate streamflow forecasting. The proposed model outperforms the streamflow persistence, ridge regression and random forest regression on majority of the gages. The model has also shown strong predictive power and can be used for long-term streamflow predictions.

Web-based distributed volunteer computing enables scientists to constitute platforms that can be used for computational tasks by using potentially millions of computers connected to the internet. The framework provides distribution and scaling capabilities for projects with user bases of thousands of volunteers. As a case study, we tested and evaluated the proposed framework with a large-scale hydrological flood forecasting model.

Citizen science opportunities for environmental monitoring have increased with the advances in smart phone capabilities and their growing availability. This project describes a new method to accurately measure river levels using smartphone sensors. Pictures of the same point on the river’s surface are taken to perform calculations based on the GPS location and spatial orientation of the smartphone. The proposed implementation is significantly more accessible than existing water measuring systems while offering similar accuracy.