Influences of the vegetation mosaic on riparian and stream environments in a mixed forest-grassland landscape in "Mediterranean" northwestern California

We examined differences in riparian and aquatic environments within the three dominant vegetation patch types of the Mattole River watershed, a 789-km(2) mixed conifer-deciduous (hardwood) forest and grassland-dominated landscape in northwestern California, USA. Riparian and aquatic environments, and particularly microclimates therein, influence the distributions of many vertebrate species, particularly the physiologically-restricted ectotherms - reptiles and amphibians (herpetofauna), and fishes. In addition to being a significant portion of the native biodiversity of a landscape, the presence and relative numbers of these more tractable small vertebrates can serve as useful metrics of its "ecological health." Our primary objective was to determine the range of available riparian and aquatic microclimatic regimes, and discern how these regimes relate to the dominant vegetations that comprise the landscape mosaic. A second objective, reported in a companion paper, was to examine relationships between available microclimatic regimes and herpetofaunal distributions. Here we examined differences in the composition, structure, and related environmental attributes of the three dominant vegetation types, both adjacent to and within the riparian corridors along 49 tributaries. Using automated dataloggers, we recorded hourly water and air temperatures and relative humidity throughout the summer at a representative subset of streams; providing us with daily means and amplitudes for these variables within riparian environments during the hottest period. Although the three vegetation types that dominate this landscape each had unique structural attributes, the overlap in plant species composition indicates that they represent a seral continuum. None-the-less, we found distinct microclimates in each type. Only riparian within late-seral forests contained summer water temperatures that could support cold-water-adapted species. We evaluated landscape-level variables to determine the best predictors of water temperature as represented by the maximum weekly maximum temperature (MWMT). The best model for predicting MWMT (adj. R-2=0.69) consisted of catchment area, aspect, and the proportion of non-forested (grassland) patches. Our model provides a useful tool for management of cold-water fauna (e.g. salmonids, stream amphibians) throughout California's "Mediterranean" climate zone.