Distribution and ecohydrologic function of pinyon pine and juniper communities and species using national longitudinal surveys
Data files
Jul 17, 2026 version files 44.79 MB
-
pj_woodlands_species-main.zip
44.79 MB
-
README.md
3.30 KB
Abstract
Pinyon (Pinus spp.) - juniper (Juniperus spp.) woodlands are a common ecosystem type across the western USA that have historically and are actively experiencing expansion and contraction with risk of ecosystem alteration in ecohydrologic function. Excluding anthropogenic influences on ecosystem development (e.g., alterations to historical fire regimes), the specific bioclimatic and topo-edaphic determinants of PJ woodland establishment, especially the species that comprise them, are poorly understood. Assessment of species-level dynamics with on-the-ground (e.g., ≤ 30 m2) indicators of ecosystem functioning is difficult with available satellite remote sensing techniques. Using ground-based plot samples, we present current (2009-2021) PJ community and species regional domains modeled from multiple longitudinal assessment and monitoring frameworks across the western USA with potential factors driving or controlling their presence. We further evaluated the ecohydrological function of PJ communities relative to pre-settlement conditions as tree densities increase and those regions where functioning may be changing. Our PJ community/ecosystem random forest classification model had an accuracy of 61.6% with species-level models ranging from 52% to 99%. The top model determinants of PJ community ecosystems were wet-season temperature, precipitation seasonality, warm-season precipitation, seasonal drought temperature, and isothermality. Total warm-season precipitation, wet-season temperature, and precipitation seasonality were among the top discriminant variables between PJ species. High PJ densities were associated with reduced species richness and total perennial grass and forb cover with consequence of increasing bare ground cover, inducing the potential for soil surface erosion (e.g., reduced ecohydrologic function). Taken together, increasing PJ species density and cover can come at the risk of loss of plant understory richness, livestock and native herbivore forage, and surface soils in absence of management intervention or fire. These national longitudinal surveys provided valuable ecosystem- and species-level detail to assess trends in both natural and agricultural resources.
This work is has been accepted:
Journal: Ecosphere
Manuscript type: Research Article
Title: Distribution and ecohydrologic function of pinyon pine and juniper woodland communities and species using national longitudinal surveys
Dataset and scripts to perform analysis and figure creation related to modeling the distribution and ecohydrologic function of pinyon pine and juniper woodlands and species.
All the R scripts and the publicly available data (Department of the Interior-Bureau of Land Management Assessment, Inventory, and Monitoring) for analysis and figure creation are provided in a single compressed/zip folder called "pj_woodlands_species-main.zip". The publicly available data was sourced from the Department of Agriculture-Agricultural Research Service's Landscape Data Commons (https://landscapedatacommons.org/). Data from the Natural Resources Conservation Service (NRCS) National Resource Inventory (NRI) are available, with restrictions, including non-disclosure agreements, licensing, confidentiality, and other agreements as mandated by law. For more details on NRCS NRI, please see the website at: https://www.nrcs.usda.gov/nri. Requests and inquiries can be directly communicated to NRCS by email at NRI@usda.gov.
R is required to run the scripts. Annotations are provided throughout the script through 1) library loading, 2) dataset loading and cleaning, 3) analyses, and 4) figure creation.
Given the number of scripts and data columns needing metadata, descriptions of individual scripts and file-level metadata are provided in separate ReadMe documents pertaining to their respective subfolder contents within pj_woodlands_species-main.zip:
pj_woodlands_species-main\PJ_Woodlands_Organize_RangeHealth_GroundCover\ReadMe.md
pj_woodlands_species-main\PJ-Woodlands_Modeling\ReadMe.md
pj_woodlands_species-main\PJ-Woodlands_Modeling\Data\ReadMe.md
Suggest working in the order listed below with the data and scripts in the folders starting with PJ_Woodlands_Organize_RangeHealth_GroundCover, then actions scripted in GEE_JavaScript_code_PJ_Extraction.txt (in Google Earth Engine), then data and scripts in the PJ-Woodlands_Modeling folder. All folders and scripts are located in pj_woodlands_species-main.zip.
Folder PJ_Woodlands_Organize_RangeHealth_GroundCover
R scripts for cleaning and organizing the AIM and NRI protocol data to include only PJ Woodland species of interest as well as Rangeland Health data ("1_data-clean-organize-PJ-Woodlands.qmd"), followed by analysis and figure creation of ground cover metrics with Rangeland Health data ("2_range_health_022024.R" scripts). This Rproj should be used first to then help with the modeling (folder PJ-Woodlands_Modeling).
File GEE_JavaScript_code_PJ_Extraction.js
JavaScript code used in Google Earth Engine for bioclimatic and topo-edaphic variable extraction by AIM and NRI point samples that was then used in the PJ Woodland modeling.
Folder PJ-Woodlands_Modeling
R scripts for random forest classification modeling the occurance of PJ Woodland ecosystem types ("1_pj_clustering_spatial_12032023.R"), followed by species level models ("pj_clustering_rf_XXX.R" scripts) and subsequent PCAs ("pj_pca_biplots_XXX.R" scripts).
