Spatial patterns of tree cover change at a dry forest margin are driven by initial conditions, water balance and wildfire

2 July 2021


Harris, L. B. and A. H. Taylor. 2021. Spatial patterns of tree cover change at a dry forest margin are driven by initial conditions, water balance and wildfire. Landscape Ecology. 36: 353-371.


Eastern slopes of the Sierra Nevada mountain range: Two sites in the central portion of the range (south of Markleeville, east of the Carson River Hwy 89) and two in the southern portion of the range (north and south of Kennedy Meadows).

Description of Site

Elevation – 1,797-2,861 m (5,896-9,386); Ppt – 422-578 mm (17-23 in); Soils – variable, but typically granitic and weakly developed; Plant communities – lower, middle, and upper elevations were mountain big sagebrush (Artemisia tridentata subsp. vaseyana) steppe, singleleaf pinyon (Pinus monophylla) and western juniper (Juniperus occidentalis), and singleleaf pinyon and Jeffrey pine (P. jeffreyi) forest, respectively. Portions of these sites burned between 2000 and 2008. Plot size – varied between 77 and 213 ha (190-526 acre).



Aerial photography was used to address questions of how terrain, water balance, pre-existing patterns of tree cover (1953), and fire influenced changes in tree cover and spatial distribution of cover over time.  Measurements – changes in tree cover at the forest-steppe ecotone was measured 1953/1955, 1999/2002, and 2016.  Tree canopy cover changes were also evaluated across elevation, slope, aspect, and topographic position index, wetness index, and preexisting tree cover.  Differences in tree cover change 1953/1955 and 2002 (prefire), and 1953/1955 to 2016 (postfire) were evaluated. Time period – 1953-2016.


There was significant infill of woodlands with >25% canopy cover (Phase II) and decline of open woodlands (Phase I) intermingling shrubland patches (<25% cover) as sites transitioned into >25% tree canopy cover.  Increases in tree canopy cover between 1953 and 2002 were positively influenced by pre-existing tree canopy cover and site-moisture.  However, wildfires in the early 2000’s resulted in net losses of tree canopy cover in 2016 compared to 1953.


The greatest increase in pre-wildfire tree canopy between 1953 and 2000 occurred in Phases I and II successional stages with Phase III stands being already fully stocked (More detailed definitions of Phases I-III available in Appendix B p. 275 in Miller et al. 2019).  Increases were also greatest in wetter more productive areas, which is supported by most of the literature.  When evaluating these results, one must remember that woodland structure and spatial pattern in 1953 was likely in a state of change since the late 1800’s.  Most woodlands across the Great Basin were experiencing infill and expansion, during the 20th Century.  An issue not addressed was invasive annuals.  Adjacent to the northern two study areas, cheatgrass (Bromus tectorum) dominates much of the understory in burned pinyon and juniper, especially in areas that burned twice between 2000 and 2010 (Robin Tausch, Retired Range Scientist, USDA Forest Service, Rocky Mountain Research Station, Reno, NV, personal communication 2021).  After 10 or more years post-fire, some of these areas have little to no tree recruitment possibly having moved into a new Steady State of invasive annual grasses.