ArcGIS REST Services Directory
JSON

Layer: Greater Sage Grouse Priority Habitat Management Area SB181D (ID:25)

View In:   Map Viewer

Name: Greater Sage Grouse Priority Habitat Management Area SB181D

Display Field: Activity_C

Type: Feature Layer

Geometry Type: esriGeometryPolygon

Description: The data set was created by preparing fine-scale population-specific Species Distribution Models (SDMs) to map revised PHMA and GHMA areas for each of the six greater sage-grouse populations within the current occupied range of Colorado. First, known presence locations of marked greater sage-grouse were used to train Random Forest and Resource Selection Function (RSF) models to estimate seasonal (e.g., breeding, summer-fall and winter) habitat suitability. Secondly, the seasonal model results were classified into high or low habitat suitability categories and subsequently compiled to produce a year-round habitat suitability map. Third, the resulting year-round habitat suitability maps were used to develop revised PHMA and GHMA areas for each population. Finally, the current occupied range for each population were modified to 1) exclude areas identified as unsuitable habitats and 2) include areas outside of current occupied range where evidence of sage-grouse occupancy exists.Data inputs into the RSF and Random Forest Models included presence data from GPS and VHF collar data provided to Olsson from CPW biologists, which was used to refine the models. A combination of vegetative and topographic predictors were employed at multiple scales in assessing the probability of habitat selection for the populations analyzed in this study. The predictors were analyzed at multiple spatial scales, as the literature demonstrates that habitat selection by a species occurs at some scales and not others (Mayor et al. 2009, Acker et al. 2017). The predictors were measured at five scales: 100 meters (m), 400 m, 1000 m, 1600 m, and 3200 m. These were selected to assess a range of local- to landscape-level scales that may influence habitat selection. Furthermore, these scales are comparable to scales assessed in other contemporary studies concerning habitat selection of greater sage-grouse (Doherty et al. 2010; Rice et al. 2016; Walker et al. 2016).Populations were also analyzed to assess utilization of smaller mapped aspen stands as compared to larger continuous forested stands of aspen and/or mixed-conifer. While greater-sage grouse tend to avoid larger forested areas, they will utilize smaller aspen stands (T. Apa pers. comm. 2016-2018). All presence locations for each population were sampled against mapped aspen stands to calculate 1) the rate of selection for aspen stands by the population, and 2) the acreage of each aspen stand utilized. The sampled stand acreages were subsequently graphed and examined to identify natural breaks in the data. Stands with acreages less than the natural break value and not directly adjacent to other forested stands were classified and analyzed separately as isolated aspen polygons which were included as potentially suitable habitat; the remaining aspen stands were classified as forested and integrated with mixed-conifer forests, which were assumed to be non-suitable habitat.Finally, the distance to forested areas was measured as a vegetative predictor using the Euclidean Distance tool in ArcGIS 10.4, excluding all isolated aspen patches and mixed-conifer patches less than 0.5 acres (and see previous paragraph).Vegetation types were derived from the Colorado Vegetation Classification Project (CVCP), a 25 m resolution raster dataset developed by CPW, which mapped landcover conditions through the periods from 1993to 1997. In addition, vegetation types were also derived from the 2001 LANDFIRE Existing Vegetation Type (EVT) layer for areas adjacent to the study area in Utah and Wyoming to provide complete and continuous vegetation cover for populations abutting the state boundary. The LANDFIRE EVT is a 30 m resolution raster dataset developed by the United States Geological Survey (USGS) mapping landcover conditions from 2001 (LANDFIRE 2001). Vegetative types were classified into biologically relevant classes and subsequently measured as percent-proportion by dividing the number of cells for the particular class by the total number of cells within the radii of the five defined scales using ArcGIS 10.4. The assigned classes of vegetative types varied by population and are detailed in the population-specific reports provided to BLM.Topographic predictors were derived from the 10 m resolution National Elevation Dataset (NED) Digital Elevation Model (DEM) developed and maintained by the USGS. Key topographic predictors include aspect, Compound Topographic Index (CTI), elevation, percent slope, slope position and surface roughness. Aspect and percent slope were calculated in ArcGIS 10.4. CTI, slope position and surface roughness were calculated using the Geomorphology and Gradient Metrics toolbox (Evans et al. 2014). In addition, aspect was subsequently transformed using the TRASP method in the Geomorphology and Gradient Metrics toolbox. To develop the multi-scale predictors, CTI and percent slope were measured as the mean of all values within the radii of the five defined scales; slope position and surface roughness were calculated using the radii of the five defined scales.The following summary of the step-wise procedure was developed to convert the Random Forest and RSF continuous surface model results into revised Habitat Management Area Prescriptions. Details of these methods follow this list:1. Classify all seasonal Random Forest and RSF model results into high and low habitat suitability layers.2. Ensemble all Random Forest and RSF classified seasonal layers to form a single year-round annual habitat layer designating locations as either high or low habitat suitability.3. Convert all highly suitable locations to Priority Habitat Management Areas (PHMA) and all locations designated as low habitat suitability to General Habitat Management Areas (GHMA).4. Classify all areas within a 0.6-mile radius from lek locations having an active or unknown status designation as PHMA, regardless of habitat suitability classification.5. Identify all irrigated agricultural lands and designate interiors as Undesignated Habitat (UDH).6. Review and apply site-specific manual conversions of initial management prescription designations based on CPW biologist and stakeholder input.7. Remove identified non-habitat areas from Current Occupied Range (COR). Expand COR in areas beyond the current population boundary where evidence exists to demonstrate occupation by greater sage-grouse.The previous habitat layer generated by CPW, only two habitat designations prescribed by the BLM ARMPA exist for assigning management approaches for conservation of the Colorado greater sage-grouse populations; PHMA and GHMA. PHMA have the highest conservation value based on a combination of habitat and sage-grouse population characteristics and are managed to minimize disturbance activities through No Surface Occupancy (NSO) stipulations and implementing capped disturbance allowances. GHMA represent areas with lower greater sage-grouse occupancy and generally have marginal habitat conditions with fewer management restrictions that provide greater flexibility in land use activities.The initial step to applying PHMA and GHMA habitat management prescriptions involves converting all areas classified as highly suitable habitat in the population’s year-round classified habitat layer to PHMA, while the remaining low habitat suitability areas are converted to GHMA. Secondly, all lek locations with a CPW-prescribed active or unknown status designation are buffered with a 0.6-mile radius and the entirety of the interior of the buffer area is converted to PHMA. Third, the most recent mapped irrigated agricultural lands data was acquired from the Colorado Division of Water Resources for all applicable populations, then the following procedure described below were implemented to apply the Undesignated Habitat prescription to the interior of all irrigated agricultural lands.Undesignated HabitatThrough the course of this study, an additional management prescription was established by AGNC to address concerns regarding habitat management on privately held irrigated agricultural lands.An Undesignated Habitat(UDH) management prescription was developed to address concerns surrounding the management of privately held irrigated agricultural lands. The UDH prescription is applicable to all populations, excluding the Parachute-Piceance-Roan population (due to a lack of irrigated agricultural lands). UDH are areas of seasonally irrigated and harvested hay fields. These areas are utilized seasonally by sage-grouse, primarily in the late summer and fall, near edges where irrigated fields are adjacent and abutting sagebrush habitats. UDH is considered effective habitat, but it is the long-term irrigation and haying practices which have created and maintain this habitat type, and thus the unimpeded irrigation, haying operations and maintenance are not considered to be a negative impact to sage-grouse. While utilization of the edges of irrigated agricultural lands by sage-grouse is known to vary from population to population, studying grouse utilization on a population-specific basis proved problematic as most populations lacked adequate telemetry locations within irrigated agricultural lands to yield results with any level of confidence. For this reason, the North Park population was selected to analyze in detail due to the high number of telemetry points located within irrigated agricultural lands. Approximately 20 percent of all summer-fall telemetry locations for the North Park population occur within irrigated agricultural lands, compared to less than 1 percent to 3 percent utilization demonstrated in the remaining populations.All summer-fall telemetry locations occurring within irrigated agricultural lands were sampled to calculate the distance each point occurred from the edges of irrigated fields. The distances for each location were plotted in a histogram and subsequently reviewed by CPW and AGNC team consultants, revealing a natural break occurring in the data at approximately 83 m. As a result, all interior irrigated agricultural lands lying beyond 83 m from the edge of sagebrush habitats are designated as UDH, while the zone occurring from the 83 meters up to the edges of sagebrush habitats retained the PHMA or GHMA designations as determined by the Random Forest and RSF model results.Final Review.Finally, the resulting revised management prescription layer was manually reviewed by AGNC and by CPW biologists and researchers to identify areas that may warrant conversion from PHMA to GHMA, or vice versa, based on biological considerations, habitat characteristics or the potential for impacting critical future economic development activities.Each population model was slightly different based on the data and nature of the populations. Below is a brief summary of the models by population along with reference to the specific population model documentation that describes each model in much more detail.Northwest Population Model: The approach to analyzing greater sage-grouse habitat suitability for the Northwest population follows the procedures outlined in section5.0 Methodsof the Methods Report. However, as described further below, the telemetry data acquired for analyzing the population were highly clustered. Employing these data in models to predict habitat suitability across a vast, variable landscape resulted in spurious and unreliable predictions in areas further removed from known presence locations. For this reason, we also developed a fourth model for the Northwest population to predict habitat suitability based on known lek locations to enhance predictions in areas lacking available telemetry data (see section 3.4.2 Random Forestbelow). Detail documentation for this model is in the Olsson Report: AGNC_GRSG_NW_Population_Report.pdfParachute-Piceance-Roan (PPR) Population Model: The approach to analyzing greater sage-grouse habitat suitability for the PPR population follows the procedures outlined in section 5.0 Methodsof the Methods Report. RSF models were not developed for the PPR population as part of this study, rather, this project employed RSF models previously developed by CPW in a separate study performed in 2015 to analyze sage-grouse habitat conditions within the PPR area (Walker et al. 2015). CPW’s analysis in their 2015 study implemented the same procedure in developing RSF models to assess habitat conditions and were therefore considered appropriate for inclusion in this project. The 2015 PPR habitat study performed by CPW differed in two respects concerning 1) scales of analyses, and 2) seasonal cut-off dates. First, CPW considered two additional scales for analyses, 800 m and 2400 m in their 2015 study, as compared to the scales analyzed in this study. To ensure consistency with CPW’s RSF models, the Random Forest models developed for this study also include these same scales of analyses. Secondly, CPW utilized marginally different cut-off dates to define seasonal date ranges. The breeding season in the 2015 PRR study was defined as occurring between March 14thto 14 days beyond the date on which approximately 90 percent of females finished nesting for that year (Walker et al. 2015). In addition, the summer-fall season was defined as ending on November 30thin CPW’s 2015 study, as compared to November 15thin this study. The telemetry data employed in Random Forest models analyzing seasonal habitat conditions in the PPR population for this study utilize the same seasonal date ranges as defined above to ensure consistency between Random Forest and RSF model analyses. CPW’s complete 2015 report, titled “Mapping and Prioritizing Seasonal Habitats for Greater Sage-Grouse in Northwestern Colorado”, and supplemental information are included as Appendix I of this report. Detail documentation for this model is in the Olsson Report: AGNC_GRSG_PPR_Population_Report.pdf.North Park Population Model: The approach to analyzing greater sage-grouse habitat suitability for the NP population follows the procedures outlined in section 5.0 Methods of the Methods Report. RSF models were not developed for the NP population as part of this study. Rather, this project employed RSF models developed by CPW in a separate study performed in 2016 to analyze sage-grouse habitat conditions for the NP range (Rice et al. 2016; Appendix I). CPW’s analysis in their 2016 study implemented similar methods to developing RSF models to assess habitat conditions and were therefore considered appropriate for inclusion in this study. The 2016 North Park habitat study performed by CPW differed in two respects concerning 1) scales of analyses, and 2) seasonal cut-off dates. First, CPW developed their seasonal RSF models for the NP population as single-scale models. The scales analyzed were derived from Average Daily Movement (ADM) distances calculated from the available telemetry data for each season. The scales employed for each seasonal analysis were 150.8 m, 83.1 m and 203.6 m for the breeding, summer-fall and winter seasons, respectively. Secondly, CPW utilized marginally different cut-off dates to define seasonal date ranges. The breeding season in the 2016 North Park study was defined as occurring between April 1st to July 15th; the summer-fall season was defined as July 16th to September 1st; the winter season was defined as starting October 1st and ending March 1st. The telemetry data employed in Random Forest models analyzing seasonal habitat conditions in the NP population for this study utilize the same seasonal date ranges as defined above to ensure consistency between Random Forest and RSF model analyses. CPW’s complete 2016 report, titled “Mapping and Prioritizing Seasonal Habitat Use by Greater Sage-Grouse (Centrocercus urophasianus) on a Landscape with Low Density Oil and Gas Development”and supplemental information are included as Appendix I of this report. Detail documentation for this model is in the Olsson Report: AGNC_GRSG_NP_Population_Report.pdf.Middle Park Population Model: The approach to analyzing greater sage-grouse habitat suitability for the Middle Park population follows the procedures outlined in Section 5.0of the Methods Report. The breeding season was defined as March 15th to June 15th, the summer-fall season was defined as June 16th to November 15th, and the winter season was defined as November 16th to March 14th. Detail documentation for this model is in the Olsson Report: AGNC_GRSG_MP_Population_Report.pdf.North Eagle-South Routt Population Model: The approach to analyzing greater sage-grouse habitat suitability for the NESR population follows the procedures outlined in section 5.0 Methodsof the Methods Report. The breeding season was defined as March 15th to June 15th, the summer-fall season was defined as June 16th to November 15th, and the winter season was defined as November 16th to March 14th. Detail documentation for this model is in the Olsson Report: AGNC_GRSG_NESR_Population_Report.pdf.Meeker White River Population Model: The approach to mapping greater sage-grouse habitat suitability for the MWR population differed from the procedures outlined in Section 5.0of the Methods Report. The population contained 97 total marked locations of greater sage-grouse across all seasons, resulting in insufficient data to adequately train habitat models to yield credible and defensible results. For this reason, this project relied on the expertise and knowledge of CPW Wildlife Managers, Moffat County officials and local landowners to manually digitize revised areas of PHMA and GHMA for the MWR population. On October 25th, 2018, CPW and AGNC representatives, as well as AGNC consultants, met with Moffat County officials and local landowners to discuss the current state of the population’s greater sage-grouse and habitat conditions, as well as the ongoing modeling efforts attempted to date. AGNC consultants and CPW informed participants that the population lacked an adequate pool of telemetry locations to perform the same modeling methods implemented on other Colorado populations in assessing habitat conditions and developing revised areas of PHMA and GHMA. For this reason, it was conveyed that the approach for revising management areas in the MWR population would rely on the local officials and landowners, as well as the input provided by the experience and expertise of CPW staff, to identify areas having the highest habitat quality evidence of utilization by greater sage-grouse. The identified areas would subsequently be digitized in a GIS and managed as PHMA, while all remaining areas would be managed according to GHMA guidelines and stipulations.In recent decades, the MWR population has seen considerable areas of lands converted from sagebrush to agricultural uses, thereby reducing the amount of suitable habitats available to the region’s sage-grouse. The conversion of these lands is most pronounced in the north-eastern lobe of the MWR population, as well as the northern portion of the western lobe of the population; the two smaller lobes in the southeast remain unaffected by agricultural operations and are predominately sagebrush habitats with smaller components of mixed-mountain shrubs.The vast majority of the lower elevations occurring in the northern portion of the western lobe of the MWR population are currently utilized in agricultural operations and host little, to no, sagebrush cover. Moving south to the higher elevations in this area of the range, the vegetation transitions to broader expanses of sagebrush intermixed with a moderate component of mixed-mountains shrubs and smaller stands of pinyon-juniper and aspen. While a considerable portion of these lands are currently protected under an assortment of Conservation Easements, these habitat conditions are nevertheless considered marginal in the ability to support greater sage-grouse occupancy. In addition, no sightings of greater sage-grouse have occurred within this region according to recent records or the available telemetry data. Finally, while three historic leks are located within this area, no evidence exists to suggest active lekking occurs within this region. In the prior decade, CPW attempted to establish two leks in this area, but the locations were not utilized and subsequently established as active lekking grounds. For these reasons, all parties agreed the western lobe of the MWR population should be managed in accordance with GHMA guidelines.The two smaller lobes located in the southeast portion of the MWR occupied range exhibit similar vegetation characteristics, though sagebrush cover is more dominant with less intermixed mixed-mountain shrubs, pinyon-juniper and aspen components. While agricultural operations are absent in these areas, there is no evidence to suggest occupancy by greater sage-grouse. There is no record of recent sightings in either area nor any marked locations of sage-grouse in the available telemetry data. Furthermore, there are no records of active, inactive or historic leks that have occurred in either lobe. Consequently, all parties concurred that both areas should be managed in accordance with GHMA guidelines.The northeastern lobe of the MWR population, located directly west of the Town of Meeker, contains the highest quality habitat within the population’s current occupied range. Despite substantial conversion of historic sagebrush lands to agricultural operations in recent decades, evidence suggests continued utilization of the surrounding habitats by greater sage-grouse, though to a far lesser degree as compared to other State populations. All 97 telemetry locations that exist for the MWR population occur in this region, consisting of three individuals collected in 2010. The population’s only active lek occurs in this area, in addition to three historic leks located on the surrounding landscape. In addition, both County officials and local landowners relayed sightings of a single individual within this area in recent years.Based on this evidence, all parties agreed that the northeastern lobe of the MWR population should be managed as both GHMA and PHMA. The northeastern lobe is bisected east to west by County Road 6. Sagebrush habitats north of County Road 6 are of lesser quality and are highly fragmented by agricultural activities, and to a lesser extent, natural gas extraction operations. Accordingly, all areas north of County Road 6 were agreed to be managed consistent with GHMA guidelines. By contrast, lands south of County Road 6 exhibit improved habitat conditions and less fragmentation. Not surprisingly, the majority of available telemetry locations occur within this region, as well as the population’s only known active lek. Therefor, it was collectively decided this region be managed in accordance with PHMA guidelines and stipulation. Furthermore, local officials and landowners indicated sightings of greater sage-grouse in this region beyond the eastern perimeter of the occupied range boundary. For this reason, the MWR boundary was expanded further east to protect additional habitats within the population’s occupied range. Detail documentation for this model is in the Olsson Report: AGNC_GRSG_MWR_Population_Report.pdf.Additional BLM and CPW Edits:The final product from Olsson was a vectorized version of the ensembled models, this resulted in a data set with a number of GIS related issues. 1) The were small grid cell sized (25 meters squared) of a particular habitat type, in a sea of another habitat type. This biologically did not make sense and was at a scale smaller than the BLM minimum mapping unit. The BLM GIS shop ran a process to remove these outlier habitat categorizations to get a more uniform layer. 2) The data were not snapped to the boundary of Colorado, the data set was intended to include range for the entire state of Colorado, however there was a sliver of non-habitat along the Wyoming and Utah border that was missing in the final version from Olsson. The CPW GIS Unit fixed this issue and snapped the Habitat layer to boundary of the state of Colorado, using the official state boundary from the State GIS data warehouse. 3) The data set was converted from a raster without the smooth option resulting in a data set with squared off boundaries at every 25 meters. This cause a large number of vertices in the data making the loading of the data set for staff very difficult. To deal with these issues the Simplify tool was used with the retain critical bends method and a simplification tolerance of 25 meters. 4) Finally, since the data was converted from a raster it no longer matched up with the boundary of the GrSG Occupied Range Layer, also there were a few omissions of occupied range that needed to be added to the final habitat layer. To rectify this the habitat layer, minus the linkage polygons was intersected with the GrSG occupied range layer and all the pieces that were not in common were saved. The pieces of the habitat layer that were outside occupied range were deleted and the pieces that were missing to complete the boundary of occupied range were added and classified as the habitat type of the adjacent polygons. These pieces were small, just triangles that caused the raster converted habitat layer to to be squared off versus the smooth occupied range layer boundary. Then small bits of occupied range that had been noticed as an omission were added.

Copyright Text: Olsson Consultants 601 P Street Suite 200 Lincoln, NE 68508 olsson.com

Min. Scale: 0

Max. Scale: 0

Default Visibility: false

Max Record Count: 4000

Supported query Formats: JSON, geoJSON, PBF

Use Standardized Queries: True

Extent:

Drawing Info:

HasZ: false

HasM: false

Has Attachments: false

Has Geometry Properties: true

HTML Popup Type: esriServerHTMLPopupTypeAsHTMLText

Object ID Field: OBJECTID

Unique ID Field:

Global ID Field:

Type ID Field:

Fields:
Templates:

Is Data Versioned: false

Has Contingent Values: false

Supports Rollback On Failure Parameter: true

Last Edit Date: 12/11/2024 8:02:45 PM

Schema Last Edit Date: 12/11/2024 8:02:45 PM

Data Last Edit Date: 12/11/2024 8:02:45 PM

Supported Operations:   Query   Query Top Features   Query Analytic   Query Bins   Generate Renderer   Validate SQL   Get Estimates   ConvertFormat