The Boulder Mountain area, including the Fremont River and Escalante Ranger Districts, is considered a top state priority focal conservation area by WRI and The Nature Conservancy. A concern of the Utah Division of Wildlife Resources is the decline of the local mule deer population, perhaps caused by the degrading condition of aspen stands. WRI and The Nature Conservancy want to take advantage of USFS proposed NEPA planning at the Fremont River Ranger District (Fishlake National Forest), and anticipated NEPA planning at the Escalante Ranger District (Dixie National Forest), to map vegetation with high resolution and accuracy, and to identify with quantitative models the most feasible of several alternative management scenarios that will achieve the greatest conservation outcomes, including the reversal of the current mule deer population decline on Boulder Mountain. Landscape Conservation Forecasting, used by The Nature Conservancy in several USFS, BLM, and NPS projects in Utah and Nevada (Low et al. 2010; Provencher et al. 2011, 2013, 2015, 2016a&b, 2017, 2018, 2019, 2021; Tuhy et al. 2010a&b, 2014), will be used to map vegetation at high spatial resolution and compare the ecological benefits of USFS NEPA alternative scenarios in FY2022-2023

Ten objectives will be completed by June 30, 2022: a) Purchase ortho-rectified Spot 6/7 1.5m resolution multispectral satellite imagery covering the southern Fremont River Ranger District and Escalante Ranger District (total of 691,249 acres). Due date: June 30, 2021 b) Describe all potential ecological systems and their reference and uncharacteristic vegetation classes in a draft Word document. Due date: June 30, 2021 c) Create Excel files containing annual/monthly time series representing future climate estimated from historic climate. Due date: June 30, 2021 d) Identify rapid-observation ground verification plots for July 2021 interpretation of unique spectral signatures on Boulder Mountains by conducting an unsupervised classification of acquired imagery. Due date: July 10, 2021. e) Conduct first field survey to correlate ground observations of Boulder Mountain ecological systems and their vegetation classes to Spot 6/7 spectral signatures. Due date: August 5, 2021. f) Deliver uninterpreted and orthorectified Spot 6/7 satellite imagery to WRI and UDWR by August 31, 2021. g) Use July field observations to draft first vegetation map and, if needed, identify rapid-observation ground verification plots for interpretation of unique spectral signatures missed in the first field survey. Due date: October 10, 2021. h) If needed, conduct second field survey to correlate ground observations of Boulder Mountain ecological systems and their vegetation classes to Spot 6/7 spectral signatures that were not surveyed in July 2021. Due date: November 10, 2021. i) Complete office interpretation of imagery using thematic classification and field observations. Due date: June 30, 2022. j) Update TNC's library of existing state-and-transition simulation models to match the new vegetation descriptions using existing models. Due date: June 30, 2022.

Risks to UDWR - Not getting high-resolution ortho-rectified satellite imagery - Not getting a description and high-resolution vegetation map of ecological systems and their vegetation classes - Not getting updated and new state-and-transition simulation models for each ecological systems of Boulder Mountain Risks to WRI - Difficulty facilitating coordination between US Forest Service, UDWR, and TNC - Description of ecological systems and vegetation map will not meet expectations of USFS for timber, northern goshawk, and at-risk species management - Satisfactory state-and-transition simulation models will not be applicable to USFS's NEPA assessment and help mitigate the decline of the mule deer population Risks to USFS - Ecological systems and vegetation classes will not reflect management needs for timber harvest and at-risk species, and lead to unsatisfactory NEPA planning - Description of conifer systems and vegetation classes could be used for potential litigation by groups that oppose timber extraction, even for at-risk species management - Vegetation map will not be useful - Wasting time and money for products that will not be used

This project will prepare data, documents, maps, and models that will inform the new NEPA assessment for the Fremont River Ranger District and prepare data for the Escalante Ranger District's when NEPA will be undertaken. The project should also inform WRI's management of the high-profiler Boulder Mountain priority focal area.

Map and modeling products will be used for timber, fuels and fire management.

In the short term there will be no effect on water quantity or quality. In the long term, results derived from data preparation could result in reduction of debris flows due to reduced likelihood of high-severity uncharacteristic large fires. Also, forest thinning that opens closed-canopy high-elevation conifer forests to about 30% canopy cover on northern slopes could result in increased runoff (Hardy et al. 2004; Ellis et al. 2013; Seyednasrollah et al. 2013).

This project requires no compliance documentation as no form of disturbance is involved. This project will however facilitate the completion of NEPA assessments and ESA Section 7 consultation. The mechanism for this facilitation lies in the quality of products that will be created and will result in the "best available science."

1. Remote sensing: a. Imagery purchase: Spot 6/7 satellite will be tasked for imagery capture dates in mid-May. Capture dates will be specified as a bracket of best dates and may include distinct dates for middle (e.g., second half of June) and high to alpine (e.g., second half of July) elevations corresponding to the period following peak plant productivity after snowmelt. Discounted archival imagery for area above 10,500 ft from 2019 or 2020 might also be acquired to allow for July field work while new imagery is being captured at the end of July or early August. Imagery will be delivered in FY2022. b. Describe ecological systems and their reference and uncharacteristic vegetation classes based on existing descriptions from TNC's 2012 Pine Valley Mountain Ranger District (Dixie National Forest) project (Tuhy et al. 2014), Indian Peak-Pine Valley-Mountain Home BLM project (Provencher et al 2021), Fremont River Ranger District Landscape Conservation Forecasting project (Tuhy 2010a), and USFS's Fishlake and Dixie national Forest 2005 Forest Plan revision modeling; however, successional classes of high-elevation conifers systems with merchatable timber will be redefined by classes that better match the classification of the USFS, if applicable, which was used to manage northern goshawk populations. TNC staff will work with USFS foresters to understand and include timber management classes. c. Identify rapid-observation ground verification plots for July 2021 interpretation of unique spectral signatures on Boulder Mountain by conducting an unsupervised classification of acquired imagery by July 10th, 2021. d. Conduct first field survey from about July 8th to 31st to correlate ground observations of Boulder Mountain ecological systems and their vegetation classes to Spot 6/7 spectral signatures. Rapid geo-referenced observations of ecological system and vegetation class will be recorded on a tablet computer and at least one georeferenced photograph of each plot observation will be taken. The area of interest will be explored in 4WD, hiking, and helicopter. e. If needed, establish rapid-observation ground verification plots for interpretation of unique spectral signatures missed in the first field survey on Boulder Mountain during October-November 2021 by conducting an unsupervised classification of unmapped areas in the acquired imagery. Task must be completed prior to October 10th, 2021. f. If needed, conduct second field survey by November 10th at the latest to correlate ground observations of Boulder Mountain ecological systems and their vegetation classes to Spot 6/7 spectral signatures that were not surveyed in July 2021. The same field protocol used during the first field survey will be used. 2. Modeling: a. Future climate time series will be obtained from PRISM data (https://prism.oregonstate.edu/). All values will be derived from annual/monthly precipitation and minimum and maximum temperature obtained from 1950 to 2020 for the middle elevation of Boulder Mountain. Future climate will be estimate from 2021 to 2081 (60 years) using the stochastic weather generator (SWG) procedure developed by Dr. Rajagopalan's civil engineering laboratory at the University of Colorado, Boulder (Verdin et al. 2014), and implemented in R and as modified by Provencher et al. (2018) for BLM's Hamlin Valley and Black Mountains project. From these times series of precipitation and min/max temperature, we will estimate the SPEI (Hayes et al. 1999) from 1950 to 2081 using the R code from Beguerí and Vicente-Serrano (2017). b. Update TNC's library of existing state-and-transition simulation models to match the new vegetation descriptions using existing models from the 2012 USFS Pine Valley Mountain Ranger District (Dixie National Forest) ST-Sim software (www.apexrms.com) database (Tuhy et al. 2014), the 2020 Indian Peak-Pine Valley-Mountain Home ST-Sim software database Provencher et al. 2021), and data files shared with TNC from USFS's 2005 Fishlake-Dixie National Forests non-spatial state-and-transition simulation models. Task will be due June 30, 2022.

The FY2021 objectives of the project are purchase of satellite imagery and Excel and Word document where monitoring may not apply. The FY2022 objective of building state-and-transition simulation model (box-and-arrow part of the models) that can be monitored for completion.

The partners are the US Forest Service Fishlake and Escalante National Forests, UDWR, and The Nature Conservancy as products will support USFS NEPA assessments and UDWR's effort to recover the Boulder Mountain mule deer population. The area of interest is also of very high conservation value to TNC.

The purpose of this project is to start an integrated phased mapping of ecological systems, assessment of forest, range, and riparian condition, and development and testing of management scenarios for on-the-ground actions in forest, shrubland, and riparian management to demonstrate where forecasted actions can be most effective for natural resources condition at the lowest cost.

The potential for increased sustainable use from this project does not apply; however, the use of the project's deliverables during next phase of the overall assessment after FY2021 has a high likelihood to produce benefits for sustainable use.

07/01/2021

08/30/2022

2023

The Nature Conservancy completed all first year deliverables of the Boulder Mountain Landscape Conservation Forecasting project. The main objectives of the first year were to a) use remote sensing to map ecological systems and their vegetation classes of Boulder Mountain, b) set-up as much as possible the spatial ST-Sim software simulation database in advance of the remote sensing contractor delivering the map, and c) prepare for the first management workshop

The following tasks were completed: 1) Mapping vegetation a) Completed a "dynamic" Word document (>35 pages) for the description of ecological systems and vegetation classes. This document is primarily for the remote sensing contractor and modelers. b) Purchase three Spot 6/7 satellite imagery captures for Boulder Mountain vegetation mapping: i) 562,907 acres captured at elevations <10,000 ft on June 11, 2021. ii) 128,342 acres captured twice at elevations >10,000 ft elevation on July 8, 2019 and August 6, 2021. We used excellent archival imagery from 2019 because TNC's field work would be completed by the end of July and remaining snow and monsoonal rains might have interfered with cloud-free and snow-free imagery capture. We tasked the satellite to image starting in mid-July, but summer rains only permitted the first decent capture on August 6th. c) The field work to associate spectral signatures (combination of red, blue, green and near infrared) and image texture to specific ecological systems and vegetation classes was conducted from 8-31 July and 18-29 October 2021. More than 5,000 observations supported with photographs were made. i) The summer field work included helicopter-based observations from 27-29 July 2021 to access remote, roadless, or rugged areas. d) Spatial Solutions delivered the vegetation map of ecological systems and vegetation classes on July 1, 2022. This version of the map contains the contractor's short-hand field codes (and not the codes in the document describing the ecological systems and vegetation classes) and would need further quality control by TNC staff and conversion from short-hand codes to 8-digit codes. 2) Simulation database set-up a) Before the Spot 6/7 imagery was captured, TNC staff created an ST-Sim simulation database initially containing 35 distinct ecological systems, The number of systems grew to 38 after map delivery. About 450 unique thematic vegetation combinations are found in the modeled ecological systems (for example, ponderosa pine mesic with a late-successional open canopy). Most ecological models existed for the same ecological systems from other western Utah and eastern Nevada areas; however, new models were created using existing templates and the ecological literature. b) TNC uploaded in ST-Sim 20 replicated climate time series that modified the variability of ecological processes (for example, fire) in all simulations. c) The delivered vegetation map that TNC cross-walked to 8-digit codes was first resampled from 1.5 m resolution to 25 m resolution to accommodate computer capacity using a resampling Python code and second split into distinct ecological system and vegetation class geotiff rasters. The two rasters were uploaded into ST-Sim in the Initial Conditions menu in addition to the land ownership raster. d) TNC created and upload in ST-Sim annual rasters (bound from 0 to 1 inclusively) describing the spatial distribution and prorated duration of use representing the deferred grazing systems used in both Ranger Districts. USFS suppled shape files of allotment and pastures, and Word documents of grazing permit per pasture/allotment that TNC converted to geotiff rasters. e) TNC also uploaded the Digital Elevation Models to control fire spread, a fire initiation probability raster based on 10-years of lightning strike data and busy roads where human fire start might start. Two rasters representing slopes from 0% to 15% and 16% to 30% to constrain use of types of mechanical equipment were created and uploaded in ST-Sim. f) After the initial assembly of the ST-Sim spatial database, TNC created a no-action (custodial management) scenario with only livestock grazing and fire-suppression and an illustrative and simplistic active management scenario with lots of prescribed fire and a limited number of mechanical canopy thinning actions. Each scenario was spatially for 25 years and replicated 20 times using climate variation. Both scenario results were ecological departure and charts of vegetation class area over 25 years. These results would be shown at the first management workshop. g) The first partner management workshop led by TNC was conducted from 27-29 September. i) The workshop resulted in map revisions. Because of these proposed revisions in progress, TNC will hold the delivery of final map layers. ii) TNC staff obtained most of the information to populate the treatment implementation menu of ST-Sim and prepare for the second management workshop (date TBD).

Recommendations for future management will be a part of the Phase 2 Boulder Mountain LCF