Sampling and modeling riparian forest structure and riparian microclimate

Abstract

Riparian areas are extremely variable and dynamic, and represent some of the most complex terrestrial ecosystems in the world. The high variability within and among riparian areas poses challenges in developing efficient sampling and modeling approaches that accurately quantify riparian forest structure and riparian microclimate. Data from eight stream reaches that are part of the Density Management Study were used in a variety of recent studies that explored sampling and modeling approaches for riparian forest structure and microclimate, and the results are summarized here. When sixteen sampling alternatives were compared based on their performance at accurately estimating the number of conifer trees per hectare, conifer basal area per hectare, and height-to diameter ratio in headwater stream reaches, rectangular strip-plots outperformed all other plot shapes. Strip-plots oriented perpendicular to the stream generally outperformed strip-plots parallel to the stream. Understory vegetation layers form a critical component of forest ecosystems. Hence, accurate estimation of their attributes (e.g., percent shrub cover) is gaining increasing importance. Percent shrub cover was modeled as a function of distance to stream and canopy leaf area index using techniques that easily accounted for spatial dependence within and among riparian areas. The distinct ecological processes, habitats, and biodiversity of riparian areas are due in part to microclimate characteristics such as air temperature (T_air) and relative humidity (RH) that differ from upland forests. Improved sampling designs and predictive models are needed to characterize riparian microclimates and their response to forest management. Height above stream and distance to stream were found to be important covariates in predicting mean maximum T_air in riparian areas. For small sample sizes, optimized sample patterns for T_air outperformed systematic sample patterns. Mean maximum T_air and mean minimum RH are strongly correlated, and mean minimum RH can be modeled as a function of mean maximum T_air and other covariates such as height above stream. Mixed eff ects models can account for within- and among-stream reach variability in RH. Application of these results can improve the quantitative estimates and reduce the costs associated with riparian forest structure and microclimate monitoring efforts.

Parent Publication

• Density management in the 21st century: west side story

Citation

Eskelson, Bianca N.I.; Anderson, Paul D.; Temesgen, Hailemariam. 2013. Sampling and modeling riparian forest structure and riparian microclimate. In: Anderson, P.D.; Ronnenberg, K.L., eds. Density management in the 21st century: west side story. Gen. Tech. Rep. PNW-GTR-880. Portland, OR: US Department of Agriculture, Forest Service, Pacific Northwest Research Station: 126–135.