“Garbage in, Garbage Out” Does Not Hold True for Indigenous Community Flood Extent Modeling in the Prairie Pothole Region

Abstract

Extensive land use changes and uncertainties arising from climate change in recent years have contributed to increased flood magnitudes in the Canadian Prairies and threatened the vulnerabilities of many small and indigenous communities. There is, thus, a need to create modernized flood risk management tools to support small and rural communities’ preparations for future extreme events. In this study, we developed spatial flood information for an indigenous community in Central Saskatchewan using LiDAR based DEM and a spatial modeling tool, the wetland DEM ponding model (WDPM). A crucial element of flood mapping in this study was community engagement in data collection, scenario description for WDPM, and flood map validation. Community feedback was also used to evaluate the utility of the modelled flood outputs. The results showed the accuracy of WDPM outputs could be improved not only with the quality of DEM but also with additional community-held information on contributing areas (watershed information). Based on community feedback, this accessible, spatially-focused modeling approach can provide relevant information for community spatial planning and developing risk management strategies. Our study found community engagement to be valuable in flood modeling and mapping by: providing necessary data, validating input data through lived experiences, and providing alternate scenarios to be used in future work. This research demonstrates the suitability and utility of LiDAR and WDPM complemented by community participation for improving flood mapping in the Prairie Pothole Region (PPR). The approach used in the study also serves as an important guide for applying transdisciplinary tools and methods for establishing good practice in research and helping build resilient communities in the Prairies.

Applying Machine Learning and Time-Series Analysis on Sentinel-1A SAR/InSAR for Characterizing Arctic Tundra Hydro-Ecological Conditions

“…Regardless, quality training data of sufficient size (i.e., number of samples) is a requirement for machine learning algorithms such as Random Forests [99,100]. Without such a dataset, acceptable classification accuracies are difficult to achieve, giving rise to the problematic computational threat of "garbage-in-and-garbageout" [101]. Our study presents a workflow that identifies the hydro-ecological state of…”

“…Regardless, quality training data of sufficient size (i.e., number of samples) is a requirement for machine learning algorithms such as Random Forests [99,100]. Without such a dataset, acceptable classification accuracies are difficult to achieve, giving rise to the problematic computational threat of "garbage-in-and-garbage-out" [101]. Our study presents a workflow that identifies the hydro-ecological state of Arctic tundra land-covers with both quantitatively and qualitatively (i.e., through visual inspection) acceptable results; however, it is recommended that future research assesses unsupervised clustering algorithms.…”

Honouring Water: The Mistawasis Nêhiyawak Water Governance Framework

“…Published by Scholarship@Western, 2022 MNFN experienced extreme flooding in 2011 and 2014 as a result of heavy snowfalls in Winter coupled with early rapid snowmelt and heavy rains in Spring. During this time, the nation experienced elevated water levels which damaged dams and levees used to prevent flooding impacts (Thapa et al, 2019). The well-being of MNFN members have been negatively impacted by contamination of water sources, deterioration of riparian habitat, road infrastructure, and displacement of people from their homes (Dawe, 2016;Thapa et al, 2019).…”

WDPM: the Wetland DEM Ponding Model

“…The program outputs have been verified by remote sensing of recent floods in this region, and WDPM has been used to develop a simpler parametric model that is more easily incorporated in hydrological models (Shook et al, 2013). The program's floodplain mapping capabilities have been used by researchers (Elboshy et al, 2019;Kiss, 2018;Schellenberg, 2017 ;Thapa et al, 2019), for operational flood hazard mapping in the Canadian Prairies by government agencies (Armstrong et al, 2013) and by private consultants (Venema, 2020a(Venema, , 2020b. As of July 2020, the program has been downloaded by at least 78 different users from 11 countries.…”