Sierra Nevada Science Symposium 2002: Science for Management and Conservation
Background BACKGROUND AND PURPOSE Conference Conveners CONVENORS Field Tours FIELD TOURS
Poster Abstracts POSTER ABSTRACTS More Information MORE INFO
Program Schedule PROGRAM SCHEDULE Mailing List MAILING LIST
Research Project Database RESEARCH PROJECTS
Location LOCATION



POSTER SESSION: Aquatic Systems/Watersheds
Abstracts for each of the posters can be viewed below by clicking on the title of the poster.



Evaporation from Lake Tahoe, California
Gayle L. Dana, Desert Research Institute, 2215 Raggio Parkway, Reno NV 89512; ph: 775-674-7538; email: gdana@dri.edu. James C. Trask, University of California, Davis. David McGraw, Desert Research Institute, Reno NV.


Accurate measurements of evaporation are important to management of water storage as well as to understanding turnover and nutrient storage in lakes. Evaporation has been a poorly constrained component of past water budget studies of Lake Tahoe, California and is the last major unknown for effective management of the Truckee River basin under the Truckee River Operating Agreement (TROA). In order to obtain good evaporation rates from Lake Tahoe two studies of evaporation were conducted from September 1999 to December 2000. The first study was designed to obtain the best possible, and first, year-round measurements of evaporation using eddy correlation technique. The second study was designed to determine the accuracy of evaporation estimates obtained from the historical Tahoe City evaporation pan, which are suspect due to progressive shading over time. In this second study, evaporation was measured with a class-A standard evaporation pan placed in a location at the U.S. Coast Guard (USCG) station a few miles from Tahoe City with minimal shading and wind obstruction.

Annual evaporation measured by eddy correlation was 672 mm, compared to 674 - 1099 mm estimated in previous studies. Annual evaporation estimated using meteorological methods (e.g., modified Penman) varied seven fold, indicating that calibration to the eddy correlation measurements would be necessary to use these methods to accurately represent annual evaporation. Pan evaporation measured at the USCG site was 1.5 - 2 times higher than at the Tahoe City site, and evaporation at both sites was higher than measured with eddy correlation. These results demonstrate that selection of a site-specific pan coefficient is essential to obtaining reliable evaporation estimates.




back to top

Wildfire Burn Patterns and Riparian Vegetation Response along Two Northern Sierra Nevada Creeks
Leda N. Kobziar* and Joe R. McBride, 145 Mulford Hall, University of California at Berkeley, Department of Environmental Science, Policy, and Management, Forest Science Division, Berkeley, CA, 94720; ph: (510) 642-4934; *email: lkobziar@nature.berkeley.edu.


Although the role of fire in forested ecosystems of the Sierra Nevada has been well-researched, little is known about how fire affects the riparian zones of these forests. In this study, we compare the effects of wildfire on riparian vegetation characteristics along two small creeks in the northern Sierra Nevada mixed-conifer forest type of the Plumas National Forest. The behavior and severity of the fire is addressed in relation to the physical characteristics of the two creeks, and the vegetation response patterns are analyzed. Where fire seemed to have burned more quickly through the transects of Fourth Water creek, more of the remaining plants sprouted than did along Third Water creek. The predominant regeneration mechanisms of the two creeks also differed. More seedlings were found on Fourth Water creek transects, while plant response along Third Water creek was mostly vegetative. For both creeks, the percent of burned hardwoods that sprouted increased with proximity to the water's edge. The interplay between fire occurrence, proximity to the water table, and how these factors influence the abundance and composition of riparian-type vegetation are explored. The results can help guide management strategies aimed at restoring resilience to the riparian corridors of disturbance-prone ecosystems.




back to top

In-Situ Overland Flow Collection System For Sierran Watersheds
C. Denton, W.W. Miller*, and D.W. Johnson, University of Nevada, Dept. Environmental & Resource Sciences, 1000 Valley Rd., Reno, NV 89512; *ph: (775) 784-4072; *email: wilymalr@ers.unr.edu. P.S.J. Verburg and G.L. Dana, Desert Research Institute, Reno, NV.


We have developed a versatile intermediate scale (50 to 100 m2 or more) in-situ overland flow collection system to capture surface runoff during natural events. Our design is easily applied to a wide range of terrain common to natural watershed settings including areas that receive considerable amounts of winter snowfall. Cumulative runoff from 13 Sierran study sites over a 7-month period ranged from a low of 3.9 L (0.08 L m-2) to a high of 21.6 L (0.43 L m-2) on slopes of <10% in areas of no visible erosion or surface runoff during both snowmelt and summer precipitation. Importantly, this approach may be useful in directly linking nutrient loading from surface runoff to tributary and lake water quality in that we found NH4+-N concentrations as high as 86.2 mg L-1 and ortho-PO4 as high as 28.7 mg L-1 P in the surface runoff. Soil solution extracts and snowmelt were > 3 and 2.5 orders of magnitude lower, respectively. These findings are highly pertinent to hydrologic and nutrient transport models in the Tahoe Basin, and likely to other watersheds having similar topographic, biological, and climatic characteristics.




back to top

How Channel and Floodplain Geometry Affect Peak Flows in High Mountain Meadows, Sierra Nevada Mountains, California
Craig Oehrli, Geologist, USDA Forest Service, Lake Tahoe Basin Management Unit, 870 Emerald Bay Road, Suite 1, South Lake Tahoe California, 96150;
ph: (530) 573-2681; email: coehrli@fs.fed.us


The purpose of this study is to determine the flood attenuation capability of meadows during spring snowmelt, winter floods, and summer thunderstorms. Most mountain meadows in the Sierra Nevada have been affected by 150 years of land use. Various land use practices have caused erosion and gully formation, stream instability, and conifer encroachment. This degradation may have diminished the meadows' natural capability to attenuate flood flows, potentially changing the magnitude, timing, and erosive power of peak flows downstream. Flow routing and geometric data from three separate meadows in the Lake Tahoe Basin combined with hypothetical flow routing will determine meadow flood attenuation capability in this fluvial geomorphic domain.

Peak inflow and outflow were recorded with an electronic current meter combined with automated stage sampling during spring snowmelt in 2000 and 2001. Topographic surveys were performed. Upcoming water surface profile and flow route modeling will determine each meadow's attenuation capability. Flood attenuation in the planned hypothetical study will be determined by routing flows through a simple compound channel. Simulations under varied flood, geometric, and antecedent characteristics will determine the approximate range of flood attenuation.

The results of this study provide water resource managers with information on flood attenuation capability of meadows under a similar fluvial geomorphic domain.




back to top

High Natural Rates of Nutrient Loading to a Montane Reservoir (Crowley Lake, CA)
D.R. Dawson* and K.N. Rose, Sierra Nevada Aquatic Research Laboratory, University of California, Santa Barbara, USA 93546; ph: (760) 935-4334; *email: dawson@icess.ucsb.edu. R. Jellison, Marine Science Institute, University of California, Santa Barbara, USA. J.M. Melack, Bren School of Environmental Science and Management, University of California, Santa Barbara, USA.


Crowley Lake, Mono County, California (area, 17 km2; mean depth, 9 m) lying in the Long Valley Caldera is a valuable aquatic resource. It is the premier trout fishery in the eastern Sierra Nevada and the largest reservoir in the Los Angeles aqueduct system. In summer, large cyanophyte blooms impair recreational uses and water quality. We measured nutrient inputs via seven tributary streams originating in alpine and subalpine catchments (total area, 985 km2) and passing through grazed lands and light urban development. On two of the tributaries, large natural spring systems contain high concentrations of nitrogen, phosphorus, and arsenic and constitute the major source of stream loading (>90% of P). Measured stream inputs of phosphorus are approximately in balance with reservoir exports, in contrast to nitrogen where exports are nearly three times the measured inputs. Thus, nitrogen fixation is a likely additional source of N.




back to top

Correlating Biological Indicators of Stress in Sierra Nevadan Lakes with Ecological Disturbance
J. Scott McClain* and Aaron Roberts, Center for Environmental Toxicology and Statistics, Miami University, Oxford, OH 45056; ph: (513) 703-3945; *email: mcclaijs@muohio.edu. Brant Allen, Tahoe Research Center, University of California, Davis. James T. Oris, Center for Environmental Toxicology and Statistics, Miami University.


Low levels of ecological degradation are difficult to observe but are important in defining alpine lakes that have low resistance and long recovery to stress. The technique of categorizing the ecological health of a lake, as applied in Eastern United States by the U.S. EPA, was applied to lakes surrounding the Lake Tahoe Basin of the Sierra Nevadan Range. Lake chemistry, littoral and riparian zone content, and human recreational activity were recorded for each of sixteen lake sites; Angora, Castle, Donner, Eagle, Fallen Leaf, Gold, Jackson Meadows, Marlette, Prosser Creek, Sand Harbor, Stampede, Spaulding, Tahoe City, Tahoe Keys, Twin, Topaz. In addition, juvenile rainbow trout (n=25) were exposed for 48 hr at each site (5 sub-sites around perimeter) in cages submerged at 2.5 m. Five genes were analyzed for mRNA levels in trout gill and liver; CYP1A1, metallothionein, vitellogenin, activin, multiple xenobiotic resistance (MXR). Gene expression analysis was statistically analyzed with principle component analysis and correlated with the most important ecological parameters of lake stress. Preliminary conclusions show a wide range of disturbance with specific chemical contact at several sites. This study suggests that inputs to lake water (run-off and motorized watercraft) can be monitored with this assessment regime and that vertebrate stress levels help to characterize local ecological disturbance.




back to top

The Effects of Prescribed Burning on Stream and Riparian Ecosystems
Leah A. Rogers* and Vincent H. Resh, 201 Wellman Hall, Division of Insect Biology, Department of Environmental Science, Policy and Management, University of California, Berkeley; ph: (510) 642-5913, *email: lrogers@nature.berkeley.edu. Scott L. Stephens, 145 Mulford Hall, Division of Forest Science, Department of Environmental Science, Policy and Management, University of California, Berkeley; ph: (510) 642-7304.


In areas where wildfire has been suppressed, prescribed burning can be an efficient forest management tool for fuel reduction and ecosystem restoration. However, concerns about the effects of fire on sensitive habitats, such as streams and riparian areas, have limited their use in management. Though wildfires (e.g., Yellowstone 1988) can have long-lasting effects on physical and biological features of streams and riparian areas, little is known about the effects of prescribed fire. In September 2002, an upland and riparian plot will be prescribed burned in the central Sierra Nevada, CA (U.C. Berkeley Blodgett Forest Research Station). We are documenting changes in water quality, channel morphology, hydrology, aquatic macroinvertebrates, algal biomass, large woody debris and riparian forest community dynamics in burned and unburned first order catchments, before and after the fire (beyond-BACI design). Multiple control and impact sites are being used to compare pre-fire and post-fire results to provide information on the: (1) effects of prescribed burning on streams and riparian zones; (2) effects of fuel reduction in riparian zones; and (3) recovery of streams following disturbance. We will present pre-fire data and results on the immediate effects of the riparian/upland prescribed fire of September 2002.




back to top

Monitoring Changes in Channel Morphology to Evaluate Management Actions Concerning the Merced River in Yosemite Valley
C. Marie Denn, PO Box 577, Yosemite, CA 95389; ph: (209) 379-1015; email: marie_denn@nps.gov.


Stream channel morphology alters according to changes in riverbank and floodplain land use, and can measurably respond to land use changes on four-to-ten year time scales. Yosemite National Park is continuing a long-term channel morphology monitoring program, initiated in 1989, to evaluate the effects of development, recreation, restoration, dam removal, and natural disturbance events on the Merced Wild and Scenic River in Yosemite Valley. This monitoring program has revealed the effects of development and intensive riverbank restoration on the river, and allowed us to better understand the 100-year flood of 1997. In the future, this study will show us alterations in the river channel structure due to post-flood policy and land us changes.




back to top

Using Ecosystem Types as Predictors of Variation in the Effects of Riparian Zones on Water Quality
Amy G. Merrill, Ph.D.; email: amerrill@nature.berkeley.edu.


The Sierra Nevada is experiencing increasingly high rates of N (nitrogen) deposition (e.g. 15 kg/ha/y) and expected increases in California's population and fossil fuel use indicate that this trend is likely to continue, possibly endangering water quality and aquatic ecosystem health. Riparian zones and other wetlands are known to remove N from ground and surface waters before they enter aquatic systems. However, variation in N filtering abilities among different riparian zones is poorly quantified, particularly in mountainous landscapes. We tested the hypothesis that different riparian ecosystem types are associated with different rates of microbial N uptake, retention, and input to ground and surface water. We randomly selected twenty plots, four each of five ecosystem types, along a tributary to Lake Tahoe. Throughout the snow free season, we measured N transformations under background and elevated N conditions. Significant differences in denitrification, net mineralization, and net nitrification were found among ecosystem types under background N conditions. Under elevated N conditions, ecosystem type differences in denitrification and ground water N flux were also significant. These results suggest that classification of riparian ecosystem types can be useful in predicting and accurately modeling landscape scale patterns of riparian zone influence on ground and stream water N.




back to top

Automated Hourly Measurements of Concentrations of Nitrate plus Nitrite and Dissolved Silica at Happy Isles Bridge, Yosemite National Park
Stephen W. Hager*, Richard E. Smith and David H. Peterson, U.S. Geological Survey, MS 496, 45 Middlefield Road, Menlo Park, CA 94025; ph: (650) 329-4587; *email: swhager@usgs.gov.


Linking variations in climate to variations in river chemistry requires monitoring chemical variations at rates similar to those of the hydro-climate variables. We are now making hourly measurements of nitrate plus nitrite (N+N) and dissolved silica (DSi) in the Merced River at Happy Isles Bridge in Yosemite National Park. Our instruments (NAS-2E and AutoLAB; W.S.Envirotech, U.K.) are user-programmable colorimetric analyzers. We have developed stable analytical routines capable of observing the small hourly variations seen in snowmelt. Precision (two standard deviations on the blank) is typically better than 0.1 micromoles per liter for N+N and 0.8 micromoles per liter for DSi. Although the NAS-2E is submergible, deployment of the analyzers on the bank of the river allows more elaborate analytical routines, larger reagent quantities, and renewable (solar) power, while ensuring the lowest possible chance of contamination of the river. Deployments of up to 12 weeks have been made with the N+N analyzer.

Data obtained demonstrate the utility of hourly sampling toward understanding the pathways the snowmelt water goes through on its way to the gage. Concentrations of N+N undergo regular diel cycles during snowmelt. At the beginning of snowmelt season, concentrations increase as flows do, the "first flush" phenomenon. Later in the year, comparison with hourly conductivity measurements reveals the relative timing of at least two pathways. Rain events following periods of dry weather also have distinct signatures.

Concentrations of DSi show a progressive decrease through the spring. Possible explanations include depletion of DSi in soils, changes in source areas, and biological uptake.




back to top

High-Resolution River Chemistry in the Sierra Nevada
David H. Peterson, U.S. Geological Survey, WRD, 345 Middlefield Road MS496, Menlo Park, CA 94025; ph: (650) 329-4525; email: dhpete@usgs.gov. R.E. Smith, S.W. Hager, M.D. Dettinger, D.R. Cayan, J.S. DiLeo, and N. K. Huber.


Linking variations in large-scale atmospheric circulation patterns (climate) to variations in snowmelt discharge and riverine chemistry is a relatively new science. Three important elements of this research are: 1) to establish a hydroclimate monitoring network; 2) to monitor riverine chemistry at rates compatible with hydroclimate variables (hourly, daily); and 3) to exploit the remarkable synchronism in spring snowmelt discharge variations between watersheds in the Sierra Nevada.

Element 3) simplifies interpreting multi-watershed variations in river chemistry. This problem is further simplified here by focusing on the variations in water conductivity, a conservative property measuring total dissolved solids (TDS) or salts (see the Hager poster for a study of the much more difficult to monitor and interpret non-conservative parameters, nitrate plus nitrite and dissolved silica concentrations).

The first (year 2001) inter basin results are from the Merced and Stanislaus Rivers. Historically, the correlation of daily discharge of these two rivers has been strong (r = + 0.98, from 1951 to 1993). (The Stanislaus discharge gage was discontinued in 1993). During 2001 a measure of discharge, water pressure or elevation, also showed a strong inter basin correlation at the hourly time scale (r = + 0.96 for calendar days 1 to 160). Despite this strong correlation in discharge, the conductivity variations in the two rivers were different. During low river flow, in both rivers, the diurnal peaks of conductivity occurred after the diurnal peak in discharge. Following the onset of snowmelt discharge, this pattern remained the same in the Stanislaus River but, in the Merced, the conductivity peak shifted, within 5 days of rising discharge, to occur before the discharge peak in the Merced. Thus, after snowmelt began, the diurnal conductivity cycles in the two rivers were 180° out of phase. A preliminary explanation for this difference is that the rate of dilution of TDS (dissolved salts) is greater in the Merced than in the Stanislaus watershed (above the gages).




back to top

Magnitude and Interannual Variability of Sediment Production from Forest Roads in the Sierra Nevada, California
Drew Coe* and Lee H. MacDonald, Department of Earth Resources, Colorado State University, Fort Collins, CO 80523; *email: drewcoe@lycos.com.


In many forested catchments roads are the primary source of sediment. However, little is known about sediment production from forest roads in the Sierra Nevada. The objectives of this study were to: (1) measure sediment production and site variables from native-surface and rocked roads, respectively; and (2) develop models to predict road surface sediment production. Sediment production rates were measured at the road segment scale by constructing 70 sediment fences and monitoring sediment production rates for 1-3 years. The road segments were in the American and Cosumnes River drainages at elevations of 900 to 2000 m, and included both public and private roads.

For the 1999-2000 wet season the mean sediment production rate for native surface roads was 7.6 t ha-1 yr-1, and the range was from 0.4 to 33 t ha-1 yr-1. For the 2000-2001 wet season the same road segments averaged only 1.4 t ha-1 yr-1, and the range of values was correspondingly reduced to 0.03 to 4.9 t ha-1 yr-1. On average, recently graded road segments produced twice as much sediment as comparable segments that had not been recently regraded. Rocked roads produced only 2-4% as much sediment as comparable native surface roads. The distribution of sediment production rates is highly skewed, as a few road segments generated most of the sediment. Preliminary data from the 2001-2002 wet season indicate that sediment production rates were intermediate to the values from the first and second wet seasons.

The large interannual variability in sediment production rates is probably due to differences in the magnitude and character of the precipitation. The more persistent snow cover in 2000-2001 appears to have protected the road surface and reduced sediment production rates per unit precipitation. Approximately 50 percent of the variability in sediment production rates for ungraded native surface roads can be explained by the product of road surface area and road gradient. The largest sediment production rates were from road segments downslope of areas with impermeable bedrock, and this may be attributed to higher runoff rates and the interception of more subsurface stormflow. Older roads with inadequate drainage produced much more sediment per unit area than roads that follow current specifications. In the absence of rocking or paving, improved drainage and road placement are the best means to reduce erosion from native surface roads.




back to top

Developing a Spatially-Explicit Model to Predict Changes in Runoff and Sediment Yields in the Central Sierra Nevada
Lee H. MacDonald*, Sandra E. Litschert, and Drew B. Coe, Department of Forest, Range, and Watershed Stewardship, Colorado State University, Fort Collins, CO 80523, ph: (970) 491-6109; *email: leemac@cnr.colostate.edu.

A lumped, conceptual model is currently being used to assess cumulative watershed effects (CWEs) on National Forest lands in California. This model converts the estimated effects of different management activities to Equivalent Roaded Acres (ERAs), sums the ERAs for the watershed of concern, and then compares the area-adjusted ERA value to an empirical threshold of concern. Key limitations to this approach include: (1) the lack of any spatial considerations (i.e., a road near the stream generally has the same ERA as a ridgetop road); (2) the absence of different coefficients and recovery rates for changes in runoff as compared to changes in erosion; and (3) the limited validation at both the site and watershed scale.

Recent increases in geographic information systems, corporate databases, computing power, and field data are facilitating the development of a spatially-explicit, quasi-physically based model to more accurately assess CWEs in the Central Sierra. The goal is to develop a modular set of procedures that allows resource specialists to rapidly predict changes in runoff and sediment yields for watersheds ranging in scale from approximately 10 to 100 km2.

In its initial phase, the model will predict catchment-scale changes in runoff and erosion due to forest harvest, roads, and fires. The lack of paired-watershed studies in the Sierra means that management-induced changes in low flows, annual water yields, and peak flows will be estimated from published values. Background and management-induced erosion rates are being obtained from a combination of literature values and field data collected by the authors from 1999-2002. The predicted changes in runoff will simply be summed over the catchment being modeled, while the sediment model will have procedures to deliver sediment from the hillslope to the channel network as a function of hillslope gradient and distance from the channel. Sediment will be routed through the stream network as a function of stream gradient, drainage area, and particle size. Users will be able to change the suggested default values for calculating changes in runoff and erosion rates, as well as altering the suggested recovery curves and algorithms for sediment delivery.

The model and user interface will be a stand-alone program in Visual Basic that uses ArcObjects and MapObjects. Standard ArcInfo coverages provide the underlying spatial data. Our design criteria are to produce a model that: (1) is relatively easy to use; (2) allows users to readily change input values and thereby evaluate the sensitivity of the model output to the assumed values; (3) can be readily expanded to accommodate other land management activities or improved predictive algorithms, and (4) will encourage the collection of additional field data by allowing users to input locally-derived values. A demonstration version of the model will be available at the conference, and a prize will be given for the best acronym for our currently unnamed model.




back to top

Geology and Landslides in Latour Demonstration State Forest
John P. Schlosser, Associate Engineering Geologist, California Geological Survey, 135 Ridgway Ave, Santa Rosa, CA 95402; ph: (707) 576-3949; email: jschloss@consrv.ca.gov. William R. Short, Senior Engineering Geologist, California Geological Survey, 1027 Tenth Street, 4th Floor, Sacramento, CA 95814; ph: (916) 322-4853; email: wshort@consrv.ca.gov. Michael A.Wopat, Senior Engineering Geologist, California Geological Survey, 6105 Airport Road, Redding, CA 96002-9422; ph: (530) 224-4748; email: mwopat@consrv.ca.gov.

The Department of Conservation's California Geological Survey (CGS) provides technical information about landslides, erosion, sedimentation, and other geologic hazards to agencies making land-use decisions in watersheds where proposed activities may affect public safety, water quality and fish habitat. At the request of the California Department of Forestry and Fire Protection, CGS conducted a study of geologic and geomorphic features related to landsliding for use in the 1994 Latour Demonstration State Forest (Shasta County) Sustained Yield Plan (SYP). The results of the study were portrayed on 1:24,000 scale maps in contiguous parts of the Miller Mountain, Hagaman Gulch, Jacks Backbone, and Manzanita Lake 7.5-minute quadrangles, which include the headwaters of South Cow Creek and Old Cow Creek. Landslide susceptibility categories were also identified. In anticipation of the 10-year update of the SYP, CGS digitized the 1994 maps using an ArcInfo Geographic Information System, with associated data attributes and metadata compiled into an ArcInfo database. Geology and landslide features on the digitized maps will be updated through field surveys, review of existing publications and air photos, and use of selected landslide-related models. Landslide potential will be recalculated so that information on both sets of maps will be consistent with methodologies currently being used by CGS for mapping landslides on California's north coast.




back to top

University of California Wildland Resources Center, UC Berkeley. Last modified: 9/27/02
©Copyright, 2001. The Regents of the University of California. For questions and comments, contact webmaster.