Weaving together ecological data with Indigenous Knowledge to model environmental factors impacting Rubus chamaemorus productivity in Southwest Alaska

Authors: KASSAMA, SIRE - Oak Ridge Institute For Science And Education (ORISE); HUNTER, GRACE - Nalaquq Llc; Friedrichsen, Claire; GLEASON, SEAN - Nalaquq Llc; Whippo, Craig; KYERE, GYABAAH - Nalaquq Llc; CHURCH, LYNN - Nalaquq Llc; Fischel, Matthew; PISARELL, KATHRYN - American Farmland Trust; BEEBE, CATHERINE - Native Village Of Kwinhagak; MATHEWS, FRANK - Native Village Of Kwinhagak; WHITE, MARGARET - Native Village Of Kwinhagak; CHURCH, MARY - Native Village Of Kwinhagak; CHURCH, WILLARD - Native Village Of Kwinhagak; MARK, DORTHY - Native Village Of Kwinhagak; MARK, JONATHON - Native Village Of Kwinhagak

Submitted to: Remote Sensing
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 6/8/2026
Publication Date: 6/11/2026
Citation: Kassama, S., Hunter, G., Friedrichsen, C.N., Gleason, S., Whippo, C.W., Kyere, G., Church, L.M., Fischel, M.H., Pisarell, K., Beebe, C., Mathews, F., White, M., Church, M., Church, W., Mark, D., Mark, J. 2026. Weaving together ecological data with Indigenous Knowledge to model environmental factors impacting Rubus chamaemorus productivity in Southwest Alaska. Remote Sensing. 18(12). https://doi.org/10.3390/rs18121939.
DOI: https://doi.org/10.3390/rs18121939

Interpretive Summary: Wild food sources are very important for the food supply, health, and culture of Indigenous communities in the Arctic. However, not many studies have used satellite images and on-the-ground checks together to track wild food plants. This study worked with the Yup’ik community in the village of Quinhagak, Alaska, to create a system for monitoring an important berry called Rubus chamaemorus (also known as atsalugpiaq or salmonberry). With help from local community members, we visited nine traditional harvesting areas to measure berry harvests. We also collected different types of data, including satellite images, elevation maps, and weather information, to build a dataset covering several years. We then used three different computer models to study how plant health, weather, and location affect berry production. The results showed that snow levels during winter, temperatures from past growing seasons, and a plant greenness index from satellite imagery were good predictors of how many berries would be produced. Overall, this study shows a method that can be used on a larger scale to track the health and productivity of important Arctic food plants, even in areas that are usually hard to study.

Technical Abstract: The spatial distribution and productivity of subsistence resources are central to food security, nutrition, and cultural vitality of circumpolar Indigenous communities. Few studies, however, have leveraged remotely sensed imagery and ground-truthing to enhance monitoring and detection of subsistence plant species. The present study uses participatory action research methodology to develop a system for monitoring the culturally important subsistence species, Rubus chamaemorus (e.g. atsalugpiaq, salmonberry) near the Alaskan Native Yup’ik village of Quinhagak in Southwest Alaska. With the assistance of Yup’ik community members, two ground-truthing surveys were conducted to assess Rubus chamaemorus productivity and species-level fractional vegetation cover at nine subsistence sites in Quinhagak’s Traditional Land Use Area (TLUA). Next, we collected PlanetScope eight-band SuperDove imagery (3m GSD); airborne lidar and satellite-derived Digital Ele-vation Models (DEMs); and four meteorological parameters to create a robust multi-year dataset. Multiple linear regression, multiple adaptive regression spline (MARS), and random forest algorithms were then tested to delineate relationships among vegetation health, climate, and spatial distribution with berry productivity. From the three model outputs, we identified dormant season snowpack, growing season temperature in previous years and the chlorophyll-related vegetation indices, MCARI, to be predictive of berry harvest outcomes. Our results demonstrate a scalable methodology for classifying the health and productivity of Arctic subsistence plants in areas that have historically posed challenging for both remote sensing and ground surveys.