Relationships between Hydrology, Vegetation and Gullies in Montane Meadows of the Southern Sierra Nevada
Hydrologist, Kings River Ranger District, Sierra National Forest
Editor's Note: This paper was not part of SNEP, but is geographically and topically relevant.
This paper presents some key findings of a three year study on montane meadows on the Sierra National Forest that was completed as a masters thesis for Humboldt State University. Sierran montane meadows have been heavily used for cattle grazing for the last 130 years, and there is accumulating evidence that meadow landforms and plant communities have changed greatly during this time. The Sierra National Forest has been stabilizing gullies in these meadows for many years and has recently overhauled the grazing program and shifted to an ecosystem approach for meadow management in order to settle a lawsuit and prevent further controversy. The biggest problem facing managers has been the gully erosion and its effects on the water table and vegetation, and there was little available information on the subject. These issues guided the objective of the thesis: to analyze the unresolved questions about meadow hydrology, vegetation and geomorphology, and to integrate this information into an understanding of the processes that form, sustain and destroy meadows.
The meadows in this study are located between 1,700 and 2,250 m in elevation and have formed in low-gradient valleys along major streamcourses in small (less than 7 km2) granitic watersheds on the Kings River Ranger District. All of the meadows in the study area have a similar hydrologic regime and can be described as "wet meadows"; such meadows are mostly saturated over most of the growing season in an average year. Similar meadows are common throughout the Sierra National Forest and in parts of Yosemite and Sequoia-Kings Canyon National Parks to the north and south, respectively. In some watersheds these meadows occupy, or once occupied, a large portion of the perennial streamcourse. Most of these meadows have been gullied to some degree, and a few have been completely excavated and de-watered.
Measuring the water budget in a study area meadow provided much insight into meadow hydrology. During the spring snowmelt, streamflow entering a meadow dominates the water budget, exceeding groundwater inflow and evapotranspirative water losses by one to three orders of magnitude. Once direct snowmelt runoff ceases in late spring, incoming streamflow drops rapidly, reflecting the shift from surface flow sources to subsurface flow sources.
In the granitic watersheds of the study area, low summer stream flows are predominately supplied by drainage from surfical colluvial deposits (forest soils) that overlie relatively impermeable bedrock. Groundwater also enters the meadow by direct flow from adjacent forested hillslopes, through the sub-soils above the bedrock, and into the meadow. These forest soils have a high water storage capacity, and are completely recharged by snowmelt even in dry years. The watersheds' groundwater flow recessions are largely controlled by the timing of snowmelt, so dry years will have an earlier peak and flows will recede sooner. The relative proportions and quantities of streamflow and groundwater entering individual meadows vary greatly and depend on the total volume of storage in the watershed. These water supplies, in combination with the amount of water used by vegetation, determine if or when the meadow water table will drop.
During the period when water tables are in the rooting zone, meadow evapotranspiration rates average 150 percent of pan evaporation, with a strong dependence on daily air temperatures. In many meadows the high evapotranspiration rates surpass inflows in the latter half of the summer. This causes the water table to drop rapidly, and streamflow leaving a meadow may fall to zero. Some meadows have an abundant supply of water and will stay saturated for the entire summer despite the high evapotranspiration rates. In late summer or fall, after a hard frost and vegetation senescence, evapotranspiration drops and the water table starts to slowly rise. However recharge will not be complete until substantial rainfall or snowmelt occurs.
One of the most interesting aspects of the montane meadows is the nature of the stream channels. Of over 100 meadows surveyed in the montane zone on the Sierra National Forest, none have the classic, entrenched gravel- or sand-bed channel that is prevalent throughout the west. The montane meadows in this part of the Sierra have a braided network of wide, shallow, completely vegetated channels that are often indistinct, particularly when the meadow vegetation is tall. During high flows the whole meadow surface conveys flow--though depths average less than 10 cm. The bankfull width-depth ratios average 42 in the naturally formed channels, and bankfull depths are less than 27 cm. Over the last one hundred years three other types of channels have developed in these meadows: (1) actively eroding "raw gullies" are characterized by a headcut, bare soil, bankfull width-depth ratios averaging 5.9 and bankfull depths greater than 30 cm; (2) "healed gullies" have a headcut but the channels are mostly vegetated and stable with bankfull width-depth ratios averaging 10.4 and minimum bankfull depths of 30 cm; and (3) "chisel channels" do not have a distinct headcut but are narrow and deep with little vegetation and are the result of chronic cattle trampling and chiseling and ensuing soil erosion. Chisel channels have bankfull width-depth ratios averaging 13.7 and bankfull depths that range from 20 to 37 cm. Gullies and chisel channels intercept and concentrate surface flows, drastically altering flow patterns.
When a gully is carved through a meadow, it intercepts and re-routes surface water and groundwater. It can cut off the water supply to a meadow aquifer downstream and leave it dry. Also, the discharge of water from gully walls steepens the hydraulic gradient, which results in a greater specific discharge and thus a lower water table. The extent to which these mechanisms alter groundwater flow patterns depends on the balance between inflows and outflows and the local hydraulic conditions. For example, a small, 30-cm-deep gully can completely intercept all surface and shallow groundwater and leave a portion of meadow with a new maximum water table that is 30 cm below the ground surface. On the other hand, if the local water supply is plentiful or converging (often the case upstream of gully heads) even a large, deep gully may have little effect in that local area. The effects of gullies on groundwater hydrology depend on the flow balance at a particular time. The effects may greatly expand during low-flow periods and affect the flow direction, velocity, and depth of groundwater flows.
In another part of the study, the plant species composition was mapped in one meadow, and the plant associations were then identified using TWINSPAN. These plant associations were compared to water table and surface flow regimes. Results show that plant associations are strongly related to hydrologic regime, and that gullies strongly affect both. At sites that have been dried by gullies, the common wet meadow species (Carex, Juncus, Eleocharis, and Moss) are replaced by mesic species (most often grasses and forbs). The results also suggest that a few months of soil saturation in each annual cycle limits where trees can establish, and that the seasonally saturated area determines the meadow-forest boundary. Lodgepole pines often invade meadows where the maximum water table has been lowered to 30 cm or more below the ground surface.
Previous stratigraphic research indicates that Sierra Nevada montane meadows have been forming over the last 10,000 years and that gully incision is a new phenomenon. Most of the meadow gully erosion in the Dinkey Creek area occurred just prior to 1940 and between 1940 and 1952--probably during the 1937 and 1950 floods. Gully head migration appears to be episodic and associated with high-magnitude floods--typical annual floods appear to be largely ineffectual. Erosion processes observed in gullies over four years indicate that subsurface processes (seepage erosion and tunnel scour) are not significant. Plunge pool erosion appears to be the dominant erosional process responsible for headcut migration in the study area meadows.
The wet meadow sods are extremely erosion resistant as they are dominated by rhizomatous species of Carex and Juncus. These sods have a minimum critical shear stress for erosion of 3000 dynes/cm2. Shear stresses generated at bankfull in natural channels average 302 dynes/cm2, with maximums around 700 dynes/cm2, which is well below the critical shear strength of the sod. Streamflow during larger floods expands over the whole meadow surface, so depth and shear stress increases are small and remain well below critical. At the downstream end of a meadow, the channels converge as the valley narrows, and the flow becomes constrained to a single, wide, vegetated channel. The flow is most concentrated here, although not sufficiently to erode a sod. Most of the large gullies originated in these constrained meadow outlets. The meadow soils below the sod are considerably more erodible, with minimum shear stresses ranging from 50 dynes/cm2 for sands to probably more than 1,000 dynes/cm2 for durable peats. These flow and shear stress conditions indicate that meadow sods had to be removed or severely damaged at several initiation points for the gullies to develop. Small patches of completely destroyed sod from cattle trampling and chiseling are ubiquitous in these meadows today, and are the most likely cause of the gullies. Flood routing indicates that relatively large and infrequent floods are required to erode durable peats, while sands would erode many times during a year. Furthermore the erosion will be most rapid and severe where small channels have been created.