Student Research Reports

View Research Report

The applications of land cover data are numerous, as land cover includes both anthropogenic and natural material, from buildings to bodies of water. Accordingly, land cover change tells the story of a location in many ways because land cover is interconnected with human and animal activity, climate and weather, and natural disasters. Our team collected photos of the land cover in eight locations around the continental United States in order to explore local land cover trends. We compared our field observation photos with several land cover datasets from remote sensing sources, identifying consistencies between datasets and our observations, as well as land cover types where disagreement occurred between sources. Our diverse locations and broad range of land cover types, coupled with additional remote sensing data related to human activity and environmental conditions, appeared well suited to an analysis of the influence of land cover The applications of land cover data are numerous, as land cover includes both anthropogenic and on wildfire risk potential. As a result of our land cover investigation, we seek to answer the question: How can we utilize artificial intelligence, remote sensing, and land cover data to classify the susceptibility of a region to wildfire spread once one begins, to assist in safety procedures and containment efforts? Our goal is to create FIRECAST: a Fire Index Risk Estimator using Climate, Anthropogenic, and Soil Trends. This exploratory tool will utilize machine learning to provide timely wildfire risk assessments to aid local communities in resource allocation and preventative measures. Key in development of FIRECAST will be our land cover photos, taken using the GLOBE Observer App, to validate land cover classification. As we continue our research, we hope to incorporate more input datasets, including more social aspects, to improve the accuracy of FIRECAST and ensure greater consistency with established fire risk indices. Keywords: land cover, GLOBE Observer, remote sensing, wildfire, risk assessment