CWHR Vegetation - Type

Vegetation maps are important for characterizing many important features of a landscape such as wildlife habitat, fuels conditions, forest composition, and carbon. Such data are most useful if they can depict vegetation type, cover, and tree size class. This version was created to capture current conditions as best as possible through a variety of existing and current sources. Cross-walks were used to compile the various sources into the common classification scheme, the California Wildlife Habitat Relationships (CWHR) system. The California Department of Forestry and Fire Protections CALFIRE Fire and Resource Assessment Program (FRAP) in cooperation with California Department of Fish and Wildlife VegCamp program and extensive use of USDA Forest Service Region 5 Remote Sensing Laboratory (RSL) (now known as Mapping and Remote Sensing Team [MARS]). has compiled the 'best available' land cover data available for California into a single comprehensive statewide data set. The data span a period from approximately 1900 to 2014. Cross-walks were used to compile the various sources into the common classification scheme, the California Wildlife Habitat Relationships (CWHR) system. There are three separate rasters provided; one for CWHR Vegetation Type, one for CWHR Tree Size Class, and one for CWHR Veg Canopy Cover (Density) Class. FVEG's WHRtype was updated with the LANDFIRE Existing Vegetation Type (EVT) data product version 2.2.0 (LANDFIRE 2020) and the Rangeland Analysis Platform (RAP) fractional ground cover data product version 3.0 (Jones et al. 2018, Allred et al. 2021). Pixels were considered for update where high severity wildfire occurred after the FVEG mapping date. high severity was defined as wildfire burned areas that experienced ≥75% loss in basal area (Parks et al. 2018, Young-Hart et al. 2022) following the wildfire event. The type of update that occurred in each 'high severity' pixel was dependent upon a lifeform conversion comparison (FVEG-to-LANDFIRE EVT) vegetation height (SALO 2020), and percent ground cover by annual and perennial grasses (RAP). Following the WHRtype update, pixels that had lifeform 'tree' then had the FVEG attributes 'WHRdensity' and 'WHRsize' updated using the SALO Forest Observatory canopy height and canopy cover data products (SALO 2020, SALO data were available for past years 2016-2020, values of canopy height and canopy cover were averaged across years for the update). To update WHRdensity, SALO canopy cover was converted to WHRdensity canopy closure class per the Wildlife Habitat Relationships, Standards for Canopy Closure Table 114C. To update WHRsize, we developed allometric equations that predict tree DBH (diameter at breast height, breast height = 4.5 ft) as a function of tree height (HT, ft). We used data from the USDA Forest Inventory and Analysis program (FIA) for California (FIA DataMart 2023, California 2022 database ver: 5.0.1). For this analysis, we included live trees ≥ 4.5 ft tall with a crown class code of dominant, co-dominant, or open grown [N = 165,224 tree measurements between 19.91 and 2019). Trees were grouped by region based on the 'fuzzed' location of the plot. Regions were defined by the original Regional Resource Kits (2023, 4 regions) and separated into softwoods and hardwoods as defined by FIA (2 categories). For each analysis, three functions were evaluated: linear, saturating, and power: Linear: DBH = a + (bHT); saturating (Michaelis-Menten) DBH = (VrmHT) /(K+HT); Power: DBH = aHTb. For the most informative model (i.e., lowest AIC), we report both the root mean squared error (RMSE) and the pseudo R1. In this case, pseudo R* was calculated as the coefficient of determination between the observed and predicted DBH. We used the most informative HT-to-DBH function for the region and tree category to convert SALO canopy height data to DBH that was then converted to WHRsize class per the Wildlife Habitat Relationships, Standards for Tree Size Table 114B.