This resource is the end product of one workshop group at the Pixels to Enviro Patterns 2026 workshop, hosted at the University of Nebraska - Lincoln.

For this project, the team used co-located NEON aquatic and terrestrial sites to tell a story of green-up, carbon-dioxide flux, and surface water availability in the Konza Prairie of Kansas, USA. We extracted one year of daily images (16:00 UTC) from GRIME-AI for two PhenoCam products at NEON D06 sites: KONA (gradient terrestrial) and KING (core aquatic). The extracted images and composite timelapse videos of the images are saved in this resource.

For KONA, we used GRIME-AI to model the Green Chromatic Coordinate (GCC) value in each image to estimate image-derived greenness over time. We then downloaded NEON's Bundled data products - eddy covariance data product (DP4.00200.001) and summarized the flux data to daily averages of CO2 flux. We made animated timeseries plot of both GCC and CO2 flux and combined the plots with the animated timelapse video from the PhenoCam to tell a story.

Additionally at KONA, we developed at Random Forest model to model CO2 flux against each output of the 'Color Segmentation' analysis tab in GRIME-AI: Entroy, GCC, GLI, NVDI, ExG, RGI. Using the model, we attempted to predict CO2 flux in various KONA images from 2024. All the information and outputs for this model are included in the PowerPoint slides in the content.

For KING, and intermittent stream, we attempted to train GRIM AI to detect the presence of water in the channel. We annotated image using SAGE as part of GRIME-AI. Using the SAGE software, we annotated 15 images of KING across different seasons and water levels to identify what is and is not water. We modeled those trained the model with the annotated set of data to produce the water detection model. Finally, we applied the water detection model to a year of images at KING, one image per day at 16:00 UTC.

This material is based in part upon work supported by the National Ecological Observatory Network (NEON), a program sponsored by the U.S. National Science Foundation (NSF) and operated under cooperative agreement by Battelle.