Spatial patterns of vigor by stand density across species groups and its drivers in a pre-harvest ponderosa pine-dominated landscape in northern California

• Published in: Forest ecology and management 2023-04, Vol.534, p.120867, Article 120867
• Main Authors: Nepal, Sushil, Eskelson, Bianca N.I., Ritchie, Martin W., Gergel, Sarah E.
• Format: Article
• Language: English
• Subjects: Abies concolor; administrative management; altitude; available water capacity; Blacks Mountain Experimental Forest; California; Calocedrus decurrens; forest ecology; forest health; forest inventory; forests; insects; landscapes; overstory; Pinus jeffreyi; Pinus ponderosa; Ponderosa pine; regression analysis; Spatial clusters; species; stand density; subsoil; Topography; vigor
• ISSN: 0378-1127; 1872-7042
• DOI: 10.1016/j.foreco.2023.120867

•High vigor, low-density ponderosa pine forests aggregated at low elevation in pre-harvest conditions.•Low vigor, high-density ponderosa pine forests aggregated at high elevation.•Low vigor, high-density incense-cedar and white fir forests aggregated at high elevation.•Spatial patterns in density by vigor and species were associated with soil moisture, soil depth, and elevation. Information on spatial patterns of forest health and stand vigor in historical forests is quite limited, particularly for forests of North America, which have been subsequently harvested. We used forest inventory data from 1934, which pre-dated early 20th century timber harvest, to reconstruct patterns of vigor in overstory trees. Across 4,000 ha of ponderosa pine-dominated forests in the Blacks Mountain Experimental Forest (BMEF), we described the spatial patterns of pre-harvest (1934) forests using a stand density index (SDI) and accounting for different species and vigor classes. We then identified topo-edaphic variables associated with pre-harvest SDI. To do so, we first calculated SDI for two species groups: 1) ponderosa pine (Pinus ponderosa Laws.) and Jeffrey pine (Pinus jeffreyi Grev. & Balf.), and 2) incense-cedar (Calocedrus decurrens (Torr.) Florin) and white fir (Abies concolor (Grod. and Glend.)) as well as for seven vigor classes within these two groups. Second, using Moran’s I, we found four spatially aggregated clusters (Moran’s I = 0.35): 1) ponderosa pine-high vigor, 2) ponderosa pine-low vigor, 3) mixed species-high vigor, and 4) mixed species-low vigor. Most of the pre-harvest landscape consisted of high vigor ponderosa pine clusters with low SDI (relative density (RD)  0.35). Using classification and regression tree analysis, we identified elevation (m), surface-soil depth (cm), and sub-soil depth (cm) were related to the clusters. A multinominal logistic regression model (kappa = 0.44; classification accuracy = 64.4 %) identified surface stoniness, available water capacity at 150 cm depth and its interaction with sub-soil depth as important variables that were related to clusters. Within ponderosa pine-dominated forests, managers consider dense stands of low vigor trees to be at-risk of insect, disease, and fire. Thus, our results can be used by managers as representative baselines for characterizing (pre-harvest) stand conditions to guide managem