Adam Collingwood
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Adam Collingwood has a total of 29 co-authors. Here are the top co-authors by number of shared publications:
A. Blaser
Department of Geography and Planning, Queen's University Kingston, ON K7L 3N6, Canada
1 shared publicationAlison Cassidy
Parks Canada, Climate and Conservation Science, Office of the Chief Ecosystem Scientist, Gatineau, Quebec, Canada
1 shared publication- Ashley C.A. Rudy
Department of Geography and Planning, Queen's University Kingston, ON K7L 3N6, Canada
1 shared publication C. Shang
Department of Geography and Planning, Queen's University Kingston, ON K7L 3N6, Canada
1 shared publication- Christina A. Braybrook
Department of Geography University of Calgary Calgary AB Canada
1 shared publication D.M. Atkinson
Department of Geography and Planning, Queen's University Kingston, ON K7L 3N6, Canada
1 shared publication
Latest Publications
- Open Access
- 1 January 2026
- Published by SPIE-Intl Soc Optical Eng in Journal of Applied Remote Sensing
- Vol. 20 (01) , 1-25
RADARSAT constellation mission and Sentinel-1 SAR data for coastal marsh mapping in Canada
Coastal marshes are important, at-risk ecosystems that require frequent monitoring to understand and quantify the effects of climate-driven change in their health and function. The use of optical Earth observation data for monitoring and mapping these ecosystems has been established, while fewer studies have assessed the benefit of synthetic aperture radar (SAR) data. Given their all-weather imaging capability and sensitivity to variations in soil moisture, surface water, vegetation structure, and biomass, SAR sensors offer multiple advantages for coastal marsh mapping. This study utilizes Sentinel-1A (S1) and RADARSAT Constellation Mission (RCM) sensors to classify coastal marshes using a random forest (RF) classifier. C-band S1 dual polarimetric 20-m resolution HH HV polarization data are classified and compared with RCM 30-m compact polarimetry data including intensity (RH, RV) along with an m-chi decomposition, which has yet to be assessed for this application. These seasonal images were acquired across three study areas within one of Canada’s largest coastal marsh complexes in Hudson Bay/James Bay Lowlands: York Factory, Manitoba; Peawanuck, Ontario; and Moosonee, Ontario. RF models to map the coastal marshes were calibrated using optically-interpreted training data, varying groupings (“datastacks”) of the SAR data as predictor variables, and then validated with ground data collected in situ at each study site. Overall accuracies of the RF classifications for the three study sites ranged from 83% to 92%. For low, mid, and high marsh sub-classes, user’s accuracies ranged from 83% to 99%, and producer’s accuracies ranged from 62% to 97%. Receiver operating characteristic plots showed that RCM m-chi decomposition polarimetric predictors resulted in higher true positive rates and lower false positive rates over all three study sites, nearing to perfect (1.0 TPR, 0.0 FPR) classifications. Next steps include extending models along the entirety of the Hudson Bay/James Bay coastline and then for all Canadian coastlines.
- 1 January 2026
- Published by SPIE-Intl Soc Optical Eng in Journal of Applied Remote Sensing
- Vol. 20 (01) , 1-25
Open AccessRADARSAT constellation mission and Sentinel-1 SAR data for coastal marsh mapping in Canada
Coastal marshes are important, at-risk ecosystems that require frequent monitoring to understand and quantify the effects of climate-driven change in their health and function. The use of optical Earth observation data for monitoring and mapping these ecosystems has been established, while fewer studies have assessed the benefit of synthetic aperture radar (SAR) data. Given their all-weather imaging capability and sensitivity to variations in soil moisture, surface water, vegetation structure, and biomass, SAR sensors offer multiple advantages for coastal marsh mapping. This study utilizes Sentinel-1A (S1) and RADARSAT Constellation Mission (RCM) sensors to classify coastal marshes using a random forest (RF) classifier. C-band S1 dual polarimetric 20-m resolution HH HV polarization data are classified and compared with RCM 30-m compact polarimetry data including intensity (RH, RV) along with an m-chi decomposition, which has yet to be assessed for this application. These seasonal images were acquired across three study areas within one of Canada’s largest coastal marsh complexes in Hudson Bay/James Bay Lowlands: York Factory, Manitoba; Peawanuck, Ontario; and Moosonee, Ontario. RF models to map the coastal marshes were calibrated using optically-interpreted training data, varying groupings (“datastacks”) of the SAR data as predictor variables, and then validated with ground data collected in situ at each study site. Overall accuracies of the RF classifications for the three study sites ranged from 83% to 92%. For low, mid, and high marsh sub-classes, user’s accuracies ranged from 83% to 99%, and producer’s accuracies ranged from 62% to 97%. Receiver operating characteristic plots showed that RCM m-chi decomposition polarimetric predictors resulted in higher true positive rates and lower false positive rates over all three study sites, nearing to perfect (1.0 TPR, 0.0 FPR) classifications. Next steps include extending models along the entirety of the Hudson Bay/James Bay coastline and then for all Canadian coastlines.
- Open Access
- 1 June 2024
- Published by Canadian Science Publishing in Arctic Science
- Vol. 10 (2) , 281-304
Remote sensing of biogeophysical variables at the Cape Bounty Arctic Watershed Observatory, Melville Island, Nunavut, Canada
The Cape Bounty Arctic Watershed Observatory (CBAWO), Melville Island, Nunavut (74°55′N, 109°34′W) was established in 2003 to examine Arctic ecosystem processes that would be impacted by climate warming and permafrost degradation. This paper provides a synthesis of how remote sensing has contributed to biogeophysical modelling and monitoring at the CBAWO from 2003 to 2023. Given the location and isolated nature of the CBAWO in the Canadian High Arctic, remote sensing data and derivatives have been instrumental for studies examining ecosystem structure and function at local and landscape scales. In combination with field measurements, remote sensing data facilitated mapping and modelling of vegetation types, % vegetation cover and aboveground phytomass, soil moisture, carbon exchange rates, and permafrost degradation and disturbance. It has been demonstrated that even in an environment with limited vegetation cover and phytomass, spectral vegetation indices (e.g., the normalized difference vegetation index) are able to model various biogeophysical variables. These applications are feasible for research sites such as the CBAWO using high spatial resolution remote sensing data across the visible, infrared, and microwave regions of the electromagnetic spectrum. Furthermore, as the satellite record continues to expand, we will gain a greater understanding of the impacts arising from the expected continued warming at northern latitudes. Although the logistics for research in the Arctic remain challenging, today's technologies (e.g., high spatial resolution satellite remote sensing, automated in situ sensors and data loggers, and wireless communication systems) can support a host of scientific endeavours in the Arctic (and other remote sites) through modelling and monitoring of biogeophysical variables and Earth surface processes with limited but critical field campaigns. The research synthesized here for the CBAWO highlights the essential role of remote sensing of terrestrial ecosystems in the Canadian Arctic.
- 1 June 2024
- Published by Canadian Science Publishing in Arctic Science
- Vol. 10 (2) , 281-304
Open AccessRemote sensing of biogeophysical variables at the Cape Bounty Arctic Watershed Observatory, Melville Island, Nunavut, Canada
The Cape Bounty Arctic Watershed Observatory (CBAWO), Melville Island, Nunavut (74°55′N, 109°34′W) was established in 2003 to examine Arctic ecosystem processes that would be impacted by climate warming and permafrost degradation. This paper provides a synthesis of how remote sensing has contributed to biogeophysical modelling and monitoring at the CBAWO from 2003 to 2023. Given the location and isolated nature of the CBAWO in the Canadian High Arctic, remote sensing data and derivatives have been instrumental for studies examining ecosystem structure and function at local and landscape scales. In combination with field measurements, remote sensing data facilitated mapping and modelling of vegetation types, % vegetation cover and aboveground phytomass, soil moisture, carbon exchange rates, and permafrost degradation and disturbance. It has been demonstrated that even in an environment with limited vegetation cover and phytomass, spectral vegetation indices (e.g., the normalized difference vegetation index) are able to model various biogeophysical variables. These applications are feasible for research sites such as the CBAWO using high spatial resolution remote sensing data across the visible, infrared, and microwave regions of the electromagnetic spectrum. Furthermore, as the satellite record continues to expand, we will gain a greater understanding of the impacts arising from the expected continued warming at northern latitudes. Although the logistics for research in the Arctic remain challenging, today's technologies (e.g., high spatial resolution satellite remote sensing, automated in situ sensors and data loggers, and wireless communication systems) can support a host of scientific endeavours in the Arctic (and other remote sites) through modelling and monitoring of biogeophysical variables and Earth surface processes with limited but critical field campaigns. The research synthesized here for the CBAWO highlights the essential role of remote sensing of terrestrial ecosystems in the Canadian Arctic.