{"id":{"repo_id":"usfca","oai_identifier":"oai:repository.usfca.edu:capstone-3464"},"canonical_url":"https://search.ndltd.org/etd/usfca/oai:repository.usfca.edu:capstone-3464","repository":{"repo_id":"usfca","name":"University of San Francisco","base_url":"https://repository.usfca.edu/do/oai/"},"display":{"title":"Complementing Topographic Wetness with Sentinel-2 Phenological Indices to Map Soil Moisture Gradients in a Snowmelt-Driven Mountain Floodplain","abstract":"Soil moisture in mountain floodplains varies over a few meters, far finer than ground-based sensors or coarse satellite images resolve; conventional optical and microwave methods sense only the top few centimeters, missing the root zone that governs plant function and biogeochemistry. Similarly, topography-derived moisture variations miss micro-heterogeneities in moisture related to intrinsic soil heterogeneity and/or the root zone. This has important implications for our ability to characterize local hotspots of biogeochemical reactions, like those related to the generation of methane. In this work, we read vegetation as a proxy for the subsurface, using Sentinel-2 land-surface phenology to infer root-zone moisture, and we test a phenology-conditioned moisture anomaly (PCMA) that reads the residual moisture signal after regressing late-season greenness on a pixel's own early-season phenology and vegetation class. Focusing our study on the upper East River watershed (Crested Butte, Colorado), we fit a mixed model of soil moisture (volumetric water content, VWC) on the wetness index, plant community, and a single Sentinel-2 phenology metric. The model was calibrated on 2018 field measurements and evaluated by leave-one-site-out cross-validation across twelve sites. Our results show that Peak NDVI carried the strongest transferable signal (out-of-site R² = 0.34, ρ = 0.53, RMSE = 7.2% VWC), out-predicting the PCMA residual. Peak NDVI ranked least important within sites (R² loss 16.3%), yet most important across sites (40.3%), and contributed most on high, steep terrain, where the wetness index showed the least predictive value. In an exploratory single-station analysis (n = 6 water years), deeper soil moisture tracked snowmelt duration (R² = 0.83) and melt rate (R² = 0.69) more closely than snowpack size (R² = 0.14). The framework maps fine-scale root-zone moisture from free inputs and flags where topography-only expectations do not suffice to capture moisture variability in mountain floodplains.","abstract_html":"Soil moisture in mountain floodplains varies over a few meters, far finer than ground-based sensors or coarse satellite images resolve; conventional optical and microwave methods sense only the top few centimeters, missing the root zone that governs plant function and biogeochemistry. Similarly, topography-derived moisture variations miss micro-heterogeneities in moisture related to intrinsic soil heterogeneity and/or the root zone. This has important implications for our ability to characterize local hotspots of biogeochemical reactions, like those related to the generation of methane. In this work, we read vegetation as a proxy for the subsurface, using Sentinel-2 land-surface phenology to infer root-zone moisture, and we test a phenology-conditioned moisture anomaly (PCMA) that reads the residual moisture signal after regressing late-season greenness on a pixel&#x27;s own early-season phenology and vegetation class. Focusing our study on the upper East River watershed (Crested Butte, Colorado), we fit a mixed model of soil moisture (volumetric water content, VWC) on the wetness index, plant community, and a single Sentinel-2 phenology metric. The model was calibrated on 2018 field measurements and evaluated by leave-one-site-out cross-validation across twelve sites. Our results show that Peak NDVI carried the strongest transferable signal (out-of-site R² = 0.34, ρ = 0.53, RMSE = 7.2% VWC), out-predicting the PCMA residual. Peak NDVI ranked least important within sites (R² loss 16.3%), yet most important across sites (40.3%), and contributed most on high, steep terrain, where the wetness index showed the least predictive value. In an exploratory single-station analysis (n = 6 water years), deeper soil moisture tracked snowmelt duration (R² = 0.83) and melt rate (R² = 0.69) more closely than snowpack size (R² = 0.14). The framework maps fine-scale root-zone moisture from free inputs and flags where topography-only expectations do not suffice to capture moisture variability in mountain floodplains.","abstract_has_math":false,"creators":["Turk, Edward"],"institution":null,"degree_name":"Master of Science in Environmental Management (MSEM)","degree_level":"Project/Capstone - Global access","degree_discipline":"Environmental Science","degree_department":null,"school":null,"contributors":["Amalia Kokkinaki"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":null,"date_issued":"","date_published":null,"updated_at":"2026-11-19T08:39:13Z","subjects":["NEON Airborne Observation Platform","Rocky Mountain Biological Laboratory","SNOTEL","lidar topographic wetness index","root-zone soil moisture remote sensing","Gunnison County Colorado","Biogeochemistry","Environmental Indicators and Impact Assessment","Environmental Monitoring"],"languages":[],"rights":[],"rights_urls":[],"identifier_entries":[],"person_identifier_entries":[]},"links":{"outbound_url":"https://repository.usfca.edu/capstone/2089","outbound_label":"Repository record","outbound_source":"dc:identifier","record_url":"https://repository.usfca.edu/capstone/2089","record_label":"Repository record","record_kind":"record","record_source":"dc:identifier","download_url":null,"download_label":null,"download_source":null,"access_status":null,"embargo_until":null,"access_note":null},"source_record":{"url":"https://repository.usfca.edu/do/oai/?verb=GetRecord&metadataPrefix=oai_etdms&identifier=oai%3Arepository.usfca.edu%3Acapstone-3464","prefix":"oai_etdms"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Amalia Kokkinaki"]},{"key":"dc:creator","label":"Author","values":["Turk, Edward"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.available","label":"Dc Date Available","values":["2026-09-01T07:00:00Z"]},{"key":"dc:date.created","label":"Dc Date Created","values":["2026-08-31T07:00:00Z"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Environmental Science"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Project/Capstone - Global access"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Master of Science in Environmental Management (MSEM)"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["NEON Airborne Observation Platform","Rocky Mountain Biological Laboratory","SNOTEL","lidar topographic wetness index","root-zone soil moisture remote sensing","Gunnison County Colorado","Biogeochemistry","Environmental Indicators and Impact Assessment","Environmental Monitoring"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://repository.usfca.edu/capstone/2089"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Soil moisture in mountain floodplains varies over a few meters, far finer than ground-based sensors or coarse satellite images resolve; conventional optical and microwave methods sense only the top few centimeters, missing the root zone that governs plant function and biogeochemistry. Similarly, topography-derived moisture variations miss micro-heterogeneities in moisture related to intrinsic soil heterogeneity and/or the root zone. This has important implications for our ability to characterize local hotspots of biogeochemical reactions, like those related to the generation of methane. In this work, we read vegetation as a proxy for the subsurface, using Sentinel-2 land-surface phenology to infer root-zone moisture, and we test a phenology-conditioned moisture anomaly (PCMA) that reads the residual moisture signal after regressing late-season greenness on a pixel's own early-season phenology and vegetation class. Focusing our study on the upper East River watershed (Crested Butte, Colorado), we fit a mixed model of soil moisture (volumetric water content, VWC) on the wetness index, plant community, and a single Sentinel-2 phenology metric. The model was calibrated on 2018 field measurements and evaluated by leave-one-site-out cross-validation across twelve sites. Our results show that Peak NDVI carried the strongest transferable signal (out-of-site R² = 0.34, ρ = 0.53, RMSE = 7.2% VWC), out-predicting the PCMA residual. Peak NDVI ranked least important within sites (R² loss 16.3%), yet most important across sites (40.3%), and contributed most on high, steep terrain, where the wetness index showed the least predictive value. In an exploratory single-station analysis (n = 6 water years), deeper soil moisture tracked snowmelt duration (R² = 0.83) and melt rate (R² = 0.69) more closely than snowpack size (R² = 0.14). The framework maps fine-scale root-zone moisture from free inputs and flags where topography-only expectations do not suffice to capture moisture variability in mountain floodplains."]},{"key":"dc:title","label":"Title","values":["Complementing Topographic Wetness with Sentinel-2 Phenological Indices to Map Soil Moisture Gradients in a Snowmelt-Driven Mountain Floodplain"]}]}],"canonical_facts":{"dc:contributor":["Amalia Kokkinaki"],"dc:creator":["Turk, Edward"],"dc:date.available":["2026-09-01T07:00:00Z"],"dc:date.created":["2026-08-31T07:00:00Z"],"dc:description.abstract":["Soil moisture in mountain floodplains varies over a few meters, far finer than ground-based sensors or coarse satellite images resolve; conventional optical and microwave methods sense only the top few centimeters, missing the root zone that governs plant function and biogeochemistry. Similarly, topography-derived moisture variations miss micro-heterogeneities in moisture related to intrinsic soil heterogeneity and/or the root zone. This has important implications for our ability to characterize local hotspots of biogeochemical reactions, like those related to the generation of methane. In this work, we read vegetation as a proxy for the subsurface, using Sentinel-2 land-surface phenology to infer root-zone moisture, and we test a phenology-conditioned moisture anomaly (PCMA) that reads the residual moisture signal after regressing late-season greenness on a pixel's own early-season phenology and vegetation class. Focusing our study on the upper East River watershed (Crested Butte, Colorado), we fit a mixed model of soil moisture (volumetric water content, VWC) on the wetness index, plant community, and a single Sentinel-2 phenology metric. The model was calibrated on 2018 field measurements and evaluated by leave-one-site-out cross-validation across twelve sites. Our results show that Peak NDVI carried the strongest transferable signal (out-of-site R² = 0.34, ρ = 0.53, RMSE = 7.2% VWC), out-predicting the PCMA residual. Peak NDVI ranked least important within sites (R² loss 16.3%), yet most important across sites (40.3%), and contributed most on high, steep terrain, where the wetness index showed the least predictive value. In an exploratory single-station analysis (n = 6 water years), deeper soil moisture tracked snowmelt duration (R² = 0.83) and melt rate (R² = 0.69) more closely than snowpack size (R² = 0.14). The framework maps fine-scale root-zone moisture from free inputs and flags where topography-only expectations do not suffice to capture moisture variability in mountain floodplains."],"dc:identifier":["https://repository.usfca.edu/capstone/2089"],"dc:subject":["NEON Airborne Observation Platform","Rocky Mountain Biological Laboratory","SNOTEL","lidar topographic wetness index","root-zone soil moisture remote sensing","Gunnison County Colorado","Biogeochemistry","Environmental Indicators and Impact Assessment","Environmental Monitoring"],"dc:title":["Complementing Topographic Wetness with Sentinel-2 Phenological Indices to Map Soil Moisture Gradients in a Snowmelt-Driven Mountain Floodplain"],"thesis:degree_discipline":["Environmental Science"],"thesis:degree_level":["Project/Capstone - Global access"],"thesis:degree_name":["Master of Science in Environmental Management (MSEM)"]},"updated_at":"2026-11-19T08:39:13Z"}