M. W. Meadows

Snowmelt timing alters shallow but not deep soil moisture in the Sierra Nevada

Roughly one-third of the Earths land surface is seasonally covered by snow. In many of these ecosystems, the spring snowpack is melting earlier due to climatic warming and atmospheric dust deposition, which could greatly modify soil water resources during the growing season. Though snowmelt timing is known to influence soil water availability during summer, there is little known about the depth of the effects and how long the effects persist. We therefore manipulated the timing of seasonal snowmelt in a high-elevation mixed-conifer forest in a Mediterranean climate during consecutive wet and dry years. The snow-all-gone (SAG) date was advanced by 6 days in the wet year and 3 days in the dry year using black sand to reduce the snow surface albedo. To maximize variation in snowmelt timing, we also postponed the SAG date by 8 days in the wet year and 16 days in the dry year using white fabric to shade the snowpack from solar radiation. We found that deeper soil water (30–60 cm) did not show a statistically significant response to snowmelt timing. Shallow soil water (0–30 cm), however, responded strongly to snowmelt timing. The drying effect of accelerated snowmelt lasted 2 months in the 0–15 cm depth and at least 4 months in the 15–30 cm depth. Therefore, the legacy of snowmelt timing on soil moisture can persist through dry periods, and continued earlier snowmelt due to climatic warming and windblown dust could reduce near-surface water storage and availability to plants and soil biota.

LiDAR‐derived snowpack data sets from mixed conifer forests across the Western United States

Airborne-based Light Detection and Ranging (LiDAR) offers the potential to measure snow depth and vegetation structure at high spatial resolution over large extents and thereby increase our ability to quantify snow water resources. Here we present airborne LiDAR data products at four Critical Zone Observatories (CZO) in the Western United States: Jemez River Basin, NM, Boulder Creek Watershed, CO, Kings River Experimental Watershed, CA, and Wolverton Basin, CA. We make publicly available snow depth data products (1 m2 resolution) derived from LiDAR with an estimated accuracy of <30 cm compared to limited in situ snow depth observations.

Evaluation of Erosion Prediction Models for Forest Roads

Forest roads can be a source of accelerated erosion, which can be detrimental to aquatic habitat, fish, and other aquatic biota. Erosion models are increasingly used to quantify sediment production from forest roads. This project evaluated the efficacy of these models to predict erosion from forest roads. Sediment production was measured from 44 road segments in the humid, temperate rain forests of Oregon and California. Sediment production from these road segments was estimated with four contemporary erosion models: the Washington Road Surface Erosion Model (WARSEM); Sediment Model 2 (SEDMODL2); WEPP:Road, an interface for the Water Erosion Prediction Project Model; and the revised universal soil loss equation (RUSLE). The erosion models consistently overestimated the amount of sediment produced by the road segments by 2 to 8 times. The results were highly variable, and there were considerable differences in erosion estimated by the four models, even for the same road segment. Further, the erosion models could not consistently identify the road segments that were the top sediment producers. It is hypothesized that the regionalized parameters used as inputs for the models do not adequately characterize the hydrology of the individual road segments. In the humid, temperate rain forests of the Pacific Northwest, surface erosion from forest roads is best predicted by the amount of runoff from the road during storms. Thus, research that will better quantify the hydrology of forest roads will provide better information to predict surface erosion from forest roads.

Hydrologic Influences On The Relationship Between Suspended Sediment Concentration and Turbidity

Establishment of total maximum daily loads (TMDLs) in water-bodies is required under the Federal Clean Water Act to meet water quality criteria. Turbidity can be used as a surrogate for suspended sediment concentrations (SSC) to predict sediment load in streams. Real time in situ measurements of turbidity were collected at a stream gauging location in the Coast Range of Oregon from October 2005-June 2006 in conjunction with the collection of individual discrete water samples. The water samples were subsequently analyzed for SSC and turbidity in a laboratory.