Data associated with "Machine learning photogrammetric analysis of images provides a scalable approach to study riverbed grain size distributions"

This data package is associated with the publication "Machine learning photogrammetric analysis of images provides a scalable approach to study riverbed grain size distribution" to be submitted to Water Resources Research (Regier et al.). The distribution of sediment grain size in streams and rivers is often quantified by the median grain size (d50), a key metric for understanding and predicting hydrologic and biogeochemical function of streams and rivers. Manual methods to measure d50 are time-consuming and ignore larger grains, while model-based methods to estimate d50 often over-generalize basin characteristics, and therefore cannot accurately represent site-scale heterogeneity. Here, we apply a machine learning photogrammetry methodology (You Only Look Once, or YOLO) for estimating d50 for grains > 2 mm based on images collected from streams and rivers throughout the Yakima River Basin (YRB). To understand how photogrammetric methods may help bridge the gaps in resolution and accuracy between manual and model-based d50 estimates, we compared YOLO d50 values to manual and model-based estimates across the YRB. We found distinct differences among methods for d50 averages and variability, and relationships between d50 estimates and basin characteristics. Source images can be found at https://data.ess-dive.lbl.gov/view/doi:10.15485/1892052. Related R scripts can be found at https://github.com/peterregier/d50_computer_vision. This dataset supports a broader study examining the drivers of spatial variability in sediment respiration rates in the Yakima River Basin. We acknowledge the Yakama Nation as owners and caretakers of the lands where we collected the data used in this project. We thank the Confederated Tribes and Bands of the Yakama Nation Tribal Council and Yakama Nation Fisheries for working with us to facilitate sample collection and optimization of data usage according to their values and worldview. This dataset is comprised of one main data folder containing (1) file-level metadata; (2) data dictionary; (3) readme; (4) d50 estimates for USGS sites within the watershed; (5) d50 estimates for images collected across the study basin; (6) digitized distribution information for d50 estimates from Abeshu et al. 2022; (7) study site characteristics; and (8) a subfolder with YOLO model results. All files are .csv, .dat, or .pdf.