Outputs for Image-based Estimation Models

This data release contains relative timeseries predictions for a single site at which a low-cost trail camera collects images at a fixed frequency. This site and others like it can be viewed at https://www.usgs.gov/apps/ecosheds/fpe/. These images collected at the site are provided, along with data annotations, to a deep learning ranking model which transforms each image into a relative quanitative estimate. This release contains one or more zip folders containing 1) the deep learning model object, 2) annotations used to train the model object, and 3) the predictions from the model object. A summary of all the data and models available at the site are contained in a comma seperated value (csv) file in this data release. The release is versioned over time as additional models or data are added for the site.

For more information about the modeling methodology, view these publications:

[1] Goodling, P.J., Fair, J.H., Gupta, A., Walker, J.D., Dubreuil, T., Hayden, M., and Letcher, B.H., 2025, Technical note: A low-cost approach to monitoring relative streamflow dynamics in small headwater streams using time lapse imagery and a deep learning model: Hydrology and Earth System Sciences, v. 29, no. 22, p. 6445–6460, at https://doi.org/10.5194/hess-29-6445-2025.

[2] Gupta, A., Chang, T., Walker, J., and B. Letcher (2022). Towards Continuous Streamflow Monitoring with Time-Lapse Cameras and Deep Learning. In ACM SIGCAS/SIGCHI Conference on Computing and Sustainable Societies (COMPASS) (COMPASS '22). Association for Computing Machinery, New York, NY, USA, 353–363. https://doi.org/10.1145/3530190.3534805

And for modeling code that can use the model objects in this data release see this software release:

[3] Walker, J.D., Gupta, A., Fair, J.B., Goodling, P.J., and Letcher, B.A., 2025, Streamflow Rank Estimation (SRE) Model: U.S. Geological Survey software release, https://doi.org/10.5066/P1YSANVM.