The Impact of Cattle Grazing in High Elevation Sierra Nevada Mountain Meadows over Widely Variable Annual Climatic Conditions ()
1. Introduction
The Sierra Nevada is a dominant land feature of the state of California, spanning 640 km (400 miles) north-tosouth and 97 to 129 km (60 to 80 miles) east-to-west [1]. With 500 peaks above 3658 m, this region’s rugged topography and other natural resources attracts 50 to 60 million recreational visitors every year [2]. Some notable Sierra landmarks include: Lake Tahoe, the largest alpine lake in lower North America; Mount Whitney, the highest peak in North America at 4418 m, and world-renowned Yosemite Valley. The mountain range provides more than half of California’s fresh surface water and acts as a natural reservoir [3,4]. Water from Sierra winter storms accumulates into snowpack, primarily between 1600 m and 4400 m. It then melts slowly through spring and into summer to fill streams and reservoirs that sustain the state’s ever-growing population and the enormous needs of lowland agriculture during the dry summer months. The Sierra Nevada’s greatest economic value comes from the abundant amount of essential fresh water the range provides to California [5].
The future dependability of this important watershed is uncertain as climate change is anticipated to increase air temperature in the Sierra Nevada causing direct impacts to the water supply [4-7]. Decreases in mean annual flow, reduced snowpack as precipitation shifts to rainfall from snowfall, and more rapid snowmelt runoff are expected with increased air temperature [4,6,8].
California’s population increased ten percent from 2000 to 2010, to nearly 38 million in 2011 [9]. As the state’s population increases, so does the demand on the limited supply of fresh water that is available [3]. Threats to Sierra Nevada watersheds include urban development, logging, mining, certain recreational uses, air pollution from the Central Valley, and summer season cattle grazing. Many experts agree that cattle grazing poses the greatest threat to water quality in undeveloped high elevation areas [10-12]. The water ecology of these areas is highly sensitive to degradation by cattle because 1) cattle cluster in stream or wetland areas, 2) there is limited or absent filtering topsoil, 3) cattle manure adds phosphates and nitrates into naturally oligotrophic bodies of water, 4) there is a short growing season, and 5) cattle cause erosion of natural stream banks, depleting the already limited top soil [13-20].
It has been well documented that cattle manure can introduce pathologic microorganisms such as Giardia, Cryptosporidium, Campylobactor, Salmonella, enterotoxic strains of Escherichia coli (E. coli), or other harmful bacteria into water [21-24]. Serious microbial water quality degradation in the Sierra Nevada, including the Stanislaus National Forest (STF), has been linked to summer cattle grazing, when manure is washed into lakes and streams or directly deposited into these bodies of water [20,25]. However, many of the earlier studies in the Sierra have been single point in time analysis, only sampling for fecal indicator bacteria after livestock exposure.
Continuous weekly analysis of stream water in cattle grazing areas both before and after the arrival of cattle would better demonstrate a cause and effect of grazing. Our research group performed such a preliminary analysis during the summer of 2009 [26]. We found a dramatic rise in indicator bacteria from a 6-week analysis before cattle arrived in alpine meadows compared with the 6-week period after grazing began. Multiple violations of state water quality standards were also found in grazed areas after cattle arrived. Our study only represented a single season, and seasonal precipitation and snowmelt varies widely in the Sierra [1,3]. Therefore the current study reported here extends our analysis of water quality over three very different climatic years.
2. Methods
Water sampling occurred exclusively in the Stanislaus National Forest (STF). The STF is located in the Sierra Nevada region of California, bordering the west and north sides of Yosemite National Park. The STF is extremely popular for outdoor recreation—fishing, hiking, camping, backpacking, swimming, rafting, canoeing, and a wide range of other outdoor recreation activities that bring more than 2,000,000 visitors each year to the forest (USDA 2009b). Within the STF itself, there are 811 miles of rivers and streams and a reservoir capacity of 768,000 acre-feet [27].
Four major watersheds are partially within the STF: the Tuolumne, Stanislaus, Mokelumne and Merced River. The Stanislaus and Tuolumne River watersheds each provide about three million acre-feet of water storage for recreation, agriculture, domestic supply, and other uses per year [28]. The Mokelumne and Merced River watersheds each provide close to a million acre-feet of water storage for recreation, agriculture, domestic supply, and other uses per year [28]. These four rivers flow into the San Joaquin River.
2.1. Field Site Selection
Livestock grazing on National Forest System lands is authorized by a grazing or livestock permit, which is issued for a ten-year term [29].
Each year our research group studied three to four grazed sites in addition to control sites where no grazing occurred. Grazed sites varied each year in accordance with which allotments were being grazed. Any given year an allotment may not be grazed as the permittee (a rancher who is issued a permit to graze livestock on Federal land) can choose not to bring cattle to the forest for summer grazing. A permittee may opt for a non-use year for a variety of reasons, such as a shortened summer grazing season due to a late spring with heavy snowpack as happened in 2010 and 2011. For example, the Long Valley/Eagle Meadow Allotment was grazed in 2009, but not grazed in 2010 and 2011.
Sample sites were selected to be representative of the several microclimates within the STF, and needed to be accessible in the six-hour holding time restraint for the bacteriological samples. Two sampling methods were used.
Method A: Collect water samples from a single location on a stream before cattle were released into the forest (the “before” samples) and during the time when cattle were present (the “after” samples).
Method B: A fence around the headwaters of the stream allowed for the “before” grazing samples and “after” grazing samples to be collected on the same date. Samples were first collected downstream of the fenced area, where cattle had access to the stream (the “outside fence/after” site). A sample was then collected on the same stream inside the fence where cattle did not have access (the “inside fence/before” site).
In 2009, four high elevation sites were sampled using Method A. Three sites were sampled in 2010 and 2011. In 2010, two sites were sampled using Method A; the third site was sampled using Method B. In 2011, one site was sampled using both Methods, and the other two sites were sampled using Method B. All of the sites are typical of grazed areas throughout the STF, and are also open and used for public recreation. All streams sampled are designated for “water contact recreation” (among other beneficial uses) by the California Regional Water Quality Control Board. One control site that was not subject to cattle grazing was also tested. The sites are described below. Table 1 provides location (i.e., latitude, longitude) coordinates for each site, using datum NAD 83. Figure 1
Table 1. List of water sample sites (lat/long datum NAD 83).
provides a vicinity area map.
2009 Sample sites:
Lower Round Meadow (site ID-LRM)—1932 m elevation Method A was used to collect water samples from a tributary stream of the Tuolumne River. Nine “before” grazing water samples and eight “after” livestock arrival water samples were collected. This site is in the Bell Meadow/Bear Lake Range Allotment.
Upper Fiddlers Green Meadow (site ID-UFG)—1966 m elevation Method A was used to collect water samples from a tributary stream of the Stanislaus River. Eight “before” grazing water samples and seven “after” livestock arrival water samples were collected. This site is in the Herring Creek Range Allotment.
Bull Run Meadow (site ID-BR)—2022 m elevation Method A was used to collect water samples from a tributary stream of the Stanislaus River. Eight “before” grazing water samples and nine “after” livestock arrival water samples were collected. This site is in the Herring Creek Range Allotment.
Barn Meadow (site ID-BM)—2273 m elevation Method A was used to collect water samples from a tributary stream of the Stanislaus River. Seven “before” grazing water samples and fifteen “after” livestock arrival water samples were collected. This site is in the Long Valley/Eagle Meadow Range Allotment.
Bourland Meadow—control site, not grazed (site IDBoM)—2225 m elevation Samples were collected from a tributary stream of the Tuolumne River. Bourland Meadow lies within a designated Research Natural Area (RNA); livestock grazing is not authorized in this area. Eight control water samples were collected during the same time that “after” samples were being collected at the other 2009 study sites.
2010 Sample Sites:
Rose Creek (RC)—1145 m elevation Method A was used to collect water samples from a tributary stream of the Stanislaus River. Three “before” water samples were and sixteen “after” livestock arrival water samples were collected. This site is within the Rushing Range Allotment.
Jawbone Creek (JC)—1733 m elevation Method A was used to collect water samples from a tributary steam of the Tuolumne River. Six “before” grazing water samples and seven “after” livestock arrival water samples were collected. This site is within the Rosasco Range Allotment.
Boggy Meadow 1 & 2 (Bog 1 & Bog 2)—Bog 1: 1694 m elevation; Bog 2: 1695 m elevation Method B was used to collect two water samples from a tributary stream of the Tuolumne River. Eight “inside fence/before” and “outside fence/after” grazing water samples were collected. This site is within the Roasaco Range Allotement.
Bourland Creek—control site, not grazed (BoM)— 2225 m elevation This site was also used as the control in 2009. Six control water samples were collected at this site during the same time that “after” samples were being collected at the other study sites in 2010.
2011 Sample Sites:
Boggy Meadow 1 & 2 (Bog 1 & Bog 2) Bog 1—1694 m elevation; Bog 2—1695 m elevation This site was also sampled in 2010 using Method B. Twelve “inside fence/before” and “outside fence/after” grazing water samples were collected in 2011.
Cottonwood Meadow (CMUS & CMUS2)—CMUS: 1733 m; CMUS2: 1767 m elevation Method B was used to collect two water samples from a tributary stream of the Tuolumne River. Five “inside fence/before” grazing water samples and six “outside fence/after” cattle arrival water samples were collected. This site is within the Rosasco Range Allotment.
Lower Round Meadow (LRM)—1932 m elevation This site was sampled in 2009 using Method A. Method A and B were used to sample this site in 2011. One “before” grazing water sample and eleven “after” livestock arrival water samples were collected from the same location. In addition, twelve “before” livestock samples were collected from an ungrazed forested area upstream of the LRM sample site.
Bourland Meadow (control site, not grazed)—2225 m elevation Control water samples were also collected at this site in 2009 and 2010. One-control water sample was collected at this site in 2011.
Additional Control Sites: The UFG, BR, and BM sample sites had “before” and “after” grazing samples collected during the summer of 2009. These sample sites were not grazed in 2011. Control water samples were collected (three at UFG and BR, four at BM) during the same time period that cattle would have been present if these areas had been grazed.
2.2. Field Water Collection
A Quality Assurance Project Plan (QAPP) was prepared for this water-monitoring project, and all procedures specified in the QAPP were followed [30].
Water samples that were collected for bacteriological testing were collected while wearing sterile gloves and collected in sample bottles sterilized and provided by a certified microbiology lab (lab) (which has Environmental Laboratory Accreditation Program [ELAP] certification). The bacteriological samples were collected before any other work was performed at the site. The sterilized Nalgene bottles hold 125 mL of liquid. They were filled to 100 mL with sample water taken directly from flowing water approximately 0.1 m below the surface. The collection date, time, and samplers’ names were recorded on the field datasheets, which are retained at the CSERC office; they are also recorded on the Chain-of-Custody form that was transmitted to the lab along with each sample. No sampling bottles were contaminated during sampling or transit.
All water samples collected for bacteriological analyses were delivered to the lab within six hours from the time the samples were collected in the field. The sample bottles were placed in Zip-loc plastic bags (to avoid potential contamination from the ice water) on ice in a cooler until delivered into the custody of the lab. While collecting the water samples, the relative flow of the stream being sampled was recorded on a field datasheet along with other observations about the sample area.
2.3. Laboratory Analyses
Water samples were delivered to a State-certified analytical laboratory in Twain Harte, CA. All water samples were tested for total coliform, fecal coliform, and E. coli bacteria using Multiple Tube Fermentation (Most Probable Number/100mL). The detection limit using this method of analysis is two-organisms/100mL of water. The detection maximum using this method of analysis is 1600-organisms/100mL of water. The analytical methods utilized by this laboratory are specified in Standard Methods for the Examination of Water and Wastewater (19th Edition).
2.4. Data Analysis
For each response variable total coliform (TC), fecal coliform (FC), and E. coli (EC) Colony Forming Unit (CFU)/100mL) the data was transformed using log (base e) to normalize the residuals. To determine the effect of year (2009, 2010, and 2011) and before grazing/after grazing/control (no grazing) on the response variables, a full factorial analysis of variance was performed (JMP IN 10, SAS Institute Inc.). This model analyzes the effect of year on the overall response variable (combined before/after/control measurements), the effect of grazing (before/after/control) on the overall response variable (combined years), and the interaction of year and grazing. For example, a significant interaction could occur if FC/100mL after grazing significantly increased from 2009 to 2010 while before and control values did not. To compare mean TC, FC and EC for the before/after effect, a Tukey’s HSD mean comparison test was performed. For measurements below (i.e., <2 Colony-Forming Units (CFU)/100mL) or above (i.e., >1600 CFU/100mL) the laboratory detection limits, the value for that sample was conservatively assumed to be equal to the limit (i.e., 2 or 1600 CFU/100mL, respectively).
The bacteria results were compared to the relevant water quality standards contained in the Central Valley Regional Water Quality Control Board’s Water Quality Control Plan for the Sacramento and San Joaquin River Basins (“Basin Plan”) [31].
In waters designated for contact recreation (REC-1), the fecal coliform concentration based on a minimum of not less than five samples for any 30-day period shall not exceed a geometric mean of 200/100ml, nor shall more than ten percent of the total number of samples taken during any 30-day period exceed 400/100ml. (Basin Plan at III-3) [31].
Data was compiled for representative 30-day periodsand results were judged as a “Type 1 Violation” whenever the geometric mean of five samples collected over a 30-day period exceeded 200 fecal coliform colonies per 100 ml of water. Results were judged as a “Type 2 Violation” whenever more than ten percent of the samples collected over a 30-day period exceeded 400 fecal coliform colonies per 100 ml of water.
2.5. Weather Data
Precipitation data was obtained from weather stations located in the Stanislaus National Forest, accessed through the California Department of Water Resources Data Exchange Center [32]. Data from three weather stations will be discussed. Two stations located within the Stanislaus River basin, one at 1707 m located near Pinecrest, the other at 2560 m located near Gianelli Meadow. The third is located within the Tuolumne River basin at 2560 m elevation located near Horse Meadow.
3. Results
The results of our study are displayed in figures 2-4. Each of the three study years produced consistent results: indicator bacteria were at very low levels before the arrival of cattle and at control sites, and increased to substantially higher levels after the arrival of cattle. The elevated levels of indicator bacteria fluctuated between samples but remained higher than before or control samples for the duration of the grazing season.
Figure 2 shows the mean level of fecal coliform before cattle were present and after the introduction of cattle by year. Each year fecal coliform increased after cattle were present. In 2009 the mean fecal coliform for the