{ "culture": "en-US", "name": "", "guid": "", "catalogPath": "", "snippet": "These data were developed to meet requirements stipulated in Oregon Senate Bill 762 (2021) and Oregon Senate Bill 80 (2023) that Oregon State University develop a map of statewide wildfire hazard and summarize results for all properties in Oregon. \nThese data represent pixel-level wildfire hazard values based on burn probability and fire intensity, and which were used to inform property-level hazard values. Data are current as of January 7, 2025.", "description": "
Wildfire hazard (hazard) in these data represent the potential for damage to structures and other human development as a result of four criteria: climate, weather, topography and vegetation. We used simulation models to quantify and map burn probability and fire intensity for all locations, and then combined burn probability and fire intensity to calculate hazard. <\/SPAN><\/SPAN><\/P> Burn probability is an estimate of the average annual likelihood that a fire will impact a given location. Pyrologix LLC. used the large fire simulator, FSim, to estimate burn probability based on 2022 landscape conditions. The modeling landscape used LANDFIRE 2.0.0 fuel data which had been updated to reflect disturbances through the end of 2021, and which was adjusted based on input from regional fire and fuels professionals. FSim uses spatial fire ignition probabilities based on historical fire records (1992 \u2013 2021) and daily weather inputs sampled from observed weather (2007 \u2013 2021) to simulate 10,000 plausible fire seasons across 14 modeling landscapes in Oregon. In each modeling landscape, Pyrologix calibrated the model using observed patterns in historical climate-fire occurrence linkages. All modeling was performed at 120-meter resolution and resampled to 30-meters. <\/SPAN><\/SPAN><\/P> Fire intensity was modeled by Pyrologix LLC using the WildEST simulation tool and the same modeling landscape described above. Pyrologix developed 216 weather scenarios based on unique combinations of wind speed, wind direction and fuel moisture sampled from observed weather records. Wildfire behavior was simulated under each weather scenario, producing fire intensity level probabilities (represented as flame lengths) based on the average flame length from simulations and the relative likelihood of each weather scenario. All modeling was performed at a 30-meter resolution. The fire intensity levels include:<\/SPAN><\/SPAN><\/P> FIL1: 0-2 ft. flame length<\/SPAN><\/P><\/LI> FIL2: 2-4 ft. flame lengths<\/SPAN><\/P><\/LI> FIL3: 4-6 ft. flame lengths<\/SPAN><\/SPAN><\/P><\/LI> FIL4: 6-8 ft. flame lengths<\/SPAN><\/SPAN><\/P><\/LI> FIL5: 8-12 ft. flame lengths<\/SPAN><\/SPAN><\/P><\/LI> FIL6: > 12 ft. flame lengths<\/SPAN><\/P><\/LI><\/OL> OAR 629-44-1026 requires that hazard be adjusted in all agricultural areas identified as irrigated in at least one of five representative years. We used IrrMapper <\/SPAN>(Ketchum et al., 2020)<\/SPAN><\/SPAN> binary data layers from 2017 \u2013 2021 and identified all pixels in Oregon that were mapped as irrigated in at least one of those layers. We clipped the IrrMapper data to the extent of mapped agricultural fields using mapped field boundaries produced by the Oregon Water Resources Department (OWRD; Bromley et al., 2024). <\/SPAN><\/SPAN><\/P> During review of draft data in spring and summer 2024, county planners from Hood River and Baker County identified data gaps in IrrMapper. OSU worked with county planners to verify the data gaps and develop corrections that were applied statewide. <\/SPAN><\/SPAN><\/P> In Hood River County, planners observed that while the majority of orchards were mapped as irrigated in IrrMapper, there were persistent holes in the data which indicated the portion of a field had not been irrigated when in fact it had. The data gaps did not appear to align with any features on the ground (e.g., recently cleared crops, change in farming practices, etc.). To address this data gap, we selected all fields shown to be growing irrigated orchard crops from the OWRD field boundaries data and added the extent of those fields to the irrigated extent identified using IrrMapper. By adding field boundary extents, we filled in data holes among orchards in IrrMapper. <\/SPAN><\/SPAN><\/P> In Baker County, planners observed that IrrMapper data frequently did not characterize wetlands as irrigated, even when the wetlands are identified as irrigated hay fields in the OWRD data. We selected freshwater forested/shrub wetlands and freshwater emergent wetlands from the National Wetland Inventory (NWI; U.S. Fish and Wildlife Service, 2023), intersected the selected wetlands with the OWRD department, and added extent to the irrigated agriculture mask. The result was a raster mask representing:<\/SPAN><\/SPAN><\/P> All areas within mapped agricultural fields identified as irrigated in at least one of five years (2017 \u2013 2021)<\/SPAN><\/SPAN><\/P><\/LI><\/UL> Irrigated orchards<\/SPAN><\/SPAN><\/P><\/LI><\/UL> Cropped wetlands<\/SPAN><\/SPAN><\/P><\/LI><\/UL> Using that irrigated agricultural mask, we adjusted burn probability and fire intensity modifier values at the corresponding locations. We decreased burn probability to 0.0001 and set the fire intensity modifier value to 10. All other burn probability and fire intensity modifier values outside the mask were not adjusted. <\/SPAN><\/SPAN><\/P> Before calculating hazard, we applied a fire intensity modifier to the fire intensity dataset. The purpose of the fire intensity modifiers is to place fire intensity on a simplified scale and to capture variation in hazard across dominant fuel types (i.e. grass, shrub, and timber). For each pixel in the fire intensity raster, based on the dominant fuel type and on which FIL the weighted average flame length was in, we substituted the following fire intensity modifiers:<\/SPAN><\/P> Grass: (FIL1) 10; (FIL2) 20; (FIL3) 30; (FIL4) 50; (FIL5) 60; (FIL6) 70<\/SPAN><\/P><\/LI> Shrub: (FIL1) 15; (FIL2) 25; (FIL3) 40; (FIL4) 60; (FIL5) 80; (FIL6) 95 <\/SPAN><\/P><\/LI> Timber: (FIL1) 20; (FIL2) 30; (FIL3) 50; (FIL4) 70; (FIL5) 80; (FIL6) 95<\/SPAN><\/P> <\/P><\/LI><\/UL> We multiplied the fire intensity modifier and burn probability datasets together at 30-meter resolution to calculate a pixel-level hazard value. Then, we averaged the pixel level hazard values within each tax lot to determine the property-level hazard value. In some cases, a tax lot did not include a pixel center and so the ArcGIS tool \u2018zonal statistics\u2019 did not report a property-level average hazard value. In those cases, we converted the tax lot to a point geometry, and extracted the underlying pixel-level hazard value. Based on average property-level pixel value, each property was assigned to one of three classes:<\/SPAN><\/SPAN><\/P> Low Hazard: 0 - < 0.001911<\/SPAN><\/P><\/LI> Moderate Hazard: >= 0.001911 - < 0.137872<\/SPAN><\/P><\/LI> High Hazard: >= 0.137872<\/SPAN><\/SPAN><\/P><\/LI><\/UL> Detailed geospatial processing methods are available in the technical guide here: <\/SPAN><\/SPAN>https://oe.oregonexplorer.info/externalcontent/wildfire/data/SB80_Public_Data.zip<\/SPAN><\/SPAN><\/A>.<\/SPAN><\/SPAN><\/P> <\/P> REFERENCES:<\/SPAN><\/SPAN><\/P> Bromley, Matt; Minor, Blake; Beamer, Jordan (2024). Oregon hydrologic area agricultural field boundaries and field level and hydrologic unit water use data [Dataset]. Dryad. <\/SPAN><\/SPAN>https://doi.org/10.5061/dryad.2v6wwpzvw<\/SPAN><\/SPAN><\/A> <\/SPAN><\/P> Ketchum, D., Jencso, K., Maneta, M.P., Melton, F., Jones, M.O., Huntington, J., 2020. IrrMapper: A Machine Learning Approach for High Resolution Mapping of Irrigated Agriculture Across the Western U.S. Remote Sensing 12, 2328. <\/SPAN><\/SPAN>https://doi.org/10.3390/rs12142328<\/SPAN><\/SPAN><\/A><\/P>