Long-term effects of tree expansion and reduction on soil climate in a semiarid ecosystem

29 July 2021


Roundy, B. A., R. F. Miller, R. J. Tausch, J. C. Chambers, and B. M. Rau. 2020.  Long-term effects of tree expansion and reduction on soil climate in a semiarid ecosystem. Ecosphere 11:e03241


Twelve SageSTEP plots distributed across the Great Basin and northern Colorado Plateau.

Description of Site

Site characteristics – soil moistures and soil temperatures ranged from mesic/aridic to frigid/xeric on generally loamy soils; Plant communities – four different woodland cover types including four western juniper (Juniperus occidentalis), four single-leaf pinyon-Utah juniper (Pinus monophyllaJ. osteosperma), two Utah juniper and two Colorado pinyon (P. edulis) sites.  Sites selected were classified as expansion woodlands into Wyoming big sagebrush (Artemisia tridentata subsp. wyomingensis) and mountain big sagebrush (A. t. subsp. vaseyana) communities (post-settlement woodlands) in Phases I, II, and III. (For complete site descriptions see Table 1 in McIver and Brunson 2014).


Treatment – control, cutting, mastication, and low to moderate prescribed fire.  Treatments resulted in an average 86% reduction in tree canopy; Time period – 12-13 years; Measurements – soil water availability and soil temperature.



Over a 13-year study period, growing degree days varied from near 30 (2018) to 100 (2016) across the SAGEsteppe plots. The number of wet degree days in spring on treated and untreated plots closely tracked October through June precipitation, rather than following a pattern of decreasing wet days with time since tree reduction. The additional wet days from tree reduction in the upper 30 cm (12 in) of soil were highest in drier years (2012, 2015 and 2018) and least in wetter years (2010, 2011, and 2016) (Fig. 4).  In the dry year of 2018 (12 to 13 years following treatment), there were 21 additional wet days in treated phase III plots compared to control.  Wet degree days generally followed the same response as wet days, which were driven more by wet days than soil temperature. The frigid/xeric regime had 19 ± 7.9 more wet days in spring than the mesic/aridic-xeric regime across the time interval of fall 2014 through spring 2018 (Table 1 and Appendix S1 Table S4 in Roundy 2020). In addition to lower soil moistures, soil temperatures were higher for the mesic/aridic-xeric than frigid/xeric regime for all seasons except winter. Burned plots also had greater cover of cheatgrass (Bromus tectorum) and annual forbs, especially for Phase III, than mechanically treated plots (19.7 vs 7.9%) (also see Williams et al. 2017 and Freund et al. 2020).


An interesting study finding was that the more water stressed the woodland, due to annual precipitation (dry or wet year), woodland phase of succession (I, II, or III), and site condition (mesic/aridic versus frigid/xeric), the greater effect tree removal had on soil water availability. In wet years, differences in available water among treatments including controls were minimal.  In dry years, there were significantly greater amounts of water available for understory plant growth compare to untreated woodlands, even 13 years after treatment. In the dry year of 2018 there were 21 more wet days in cut phase III compared to uncut woodlands. The magnitude of increase in available soil water (not total) was greater on drier sites (mesic/aridic) and sites in the late stage of woodland succession (phase III) following tree reduction. A similar pattern occurred in soil water availability between cut and uncut woodlands between wet and dry years in an earlier study in western juniper communities (see Figs. 1 and 2 in Bates et al. 2000).  The greater magnitude of increased soil water on dry sites and sites in phase III, not only benefits desirable understory species but can lower the resistance of the site to invasive species.