James W. Bartolome

SAGEBRUSH-GRASS VEGETATION DYNAMICS: COMPARING CLASSICAL AND STATE-TRANSITION MODELS

The State-Transition (ST) model has been proposed as a replacement for the widely used Classical linear succession model and its derivative Range Condition (RC) model for describing and predicting rangeland community dynamics in response to man- agement. Although the ST model offers significant advantages because it accommodates nonlinear and nonequilibrium theory and is more amenable to quantitative testing of hy- potheses about community change, to date its applications have not fully utilized those advantages using empirical data. We compare the utility of the Classical, RC, and ST models in the Artemisia tridentatalPseudoroegneria spicata vegetation type, utilizing a long-term data set from southeastern Oregon. First we develop and examine the Classical and RC models for their ability to describe and predict observed vegetation changes; second we develop an ST model by classifying and quantitatively identifying states and transitions that were observed over a period of 20 yr. The Classical and RC models adequately describe most of the observed changes in vegetation through use and application of broad descriptive categories on extensive range sites. This greatly reduces the utility and predictive value of the Classical and RC approaches as guides to management and restoration. The states and transitions developed quantitatively for the ST model offer considerably more precision and predictive value than the Classical/RC models but require large, long-term, site-specific data sets, which are usually not available for rangeland systems. The specific seral stages developed for the Classical model and commonly described in the literature differ signif- icantly from the states and transitions derived quantitatively from the empirical data. States in the ST model exhibit a significant time dependency and could not have been adequately developed without long-term observations.

A state-transition approach to understanding nonequilibrium plant community dynamics in Californian grasslands

Usinga spatially and temporally replicated dataset, we built astate-transition model for Californian grasslands. We delineatedvegetation states by allowing TWINSPAN to classifyplot-level (≈10 m2) species cover datacollected over 3 to 5 consecutive years on 9 sites in an experimental designthat incorporated 5 residual dry matter (RDM) treatment levels representativeofthe range of grazing management prescriptions for this type (0, 280, 560, 841,1121 kg RDM·ha−1). We identified anddescribed a new California annual grassland subtype – Coast RangeGrassland – that is distinct from the previously described Coastal Prairieand Valley Grassland. Classification and regression tree(CART) analysis correctly classified 63% ofTWINSPAN-created vegetation transitions amongstates with interactions among site and monthly climate averages as the maindriving factors. The RDM variable (a surrogate for grazing intensity) wasimportant in model refinement, but only at a few site × year combinations andpredictions were rarely attributable to the grazing intensity gradient. Theequilibrium-based conclusion that grazing intensity manipulation createsdistinctive community structure was restricted in application to a few sites.The results suggest that equilibrium models may be appropriate for predictingsystem productivity but not the community composition, details of which requirea nonequilibrium approach. The nonequilibrium state-transition modeloffers considerable potential for improving the development and testing ofhypotheses about vegetation change and the limitations of management controls,but will require relatively large spatially and temporally replicated datasets.

Influence of natural mulch on forage production on differing California annual range sites.

Manipulation of natural mulch on nine experimental plots in California annual grassland representing a range of mean annual precipitation from I60 to 16 cm provided information useful for grazing management. Peak standing crop correlated highly significantly with precipitation. Response of peak standing crop to five levels of natural mulch ranging from zero to 1,120 kg/ha differed with site. Three types of sites distinguished by mean annual precipitation and plant species composition were identified. On sites with significant numbers of perennial grasses and more than 150 cm of mean annual precipitation, maximum standing crop is reached when more than 1,120 kg/ha of mulch is present on the ground at the beginning of the fall growing season. Peak standing crop results from 840 kg/ha of mulch on sites containing the annuals Bromus mollis and Erodium botrys and with between 100 and 65 cm of mean annual precipitation. Mulch did not significantly influence standing crop in regions dominated by Bromus rubens and Erodium cicutarium and receiving less than 25 cm of mean annual precipitation. Annual grassland response to mulch and grazing is highly site specific, yet the resilience of annual rangelands also allows rapid recovery from overuse. A distinctive combination of use by people and the influence of geography and climate fostered the formation of the present California annual grassland. The vegetation is dominated primarily by annual grasses of Mediterranean origin, which replaced the native bunchgrasses during the 19th century. Reflecting the diverse geography of the state, the California annual grassland varies in productivity, component species, and the degree of replacement of the native plants by aliens. The climate, although generally typically Mediterrean with cool wet winters and dry summers, varies for example, in average annual rainfall, from nearly 200 cm (80 inches) in the North Coast to less than 15 cm (6 inches) in the rainshadow of the South Coast ranges. The growth of vegetation follows a characteristic pattern. In response to fall rains, plants establish from seed produced the previous spring. Slow vegetative growth and root development in the cool winter months are followed by rapid growth with warming temperatures in spring. Maximum annual standing crop generally occurs in late spring. In summer only the seeds are left as living agents to repeat this process again in the fall. The authors are lecturer and associate specialist, Department of Forestry and Resource Management, University of California, Berkeley, 94720; conservationist (civilian), Department of the Navy, Western Div., Naval Facilities Engineering Command, San Bruno, California 94066; and assistant vice president, Division of Agricultural Sciences, University of California, Berkeley 94720. Glen D. Savelle established and experimental plots and carried out the initial treatments and sampling. Michael D. Pitt contributed significantly to later sampling and plot manipulation. The cooperation of private and public landholders is gratefully acknowledged. Manuscript received November 20, 1978. 4 Much of California’s range livestock industry relies on the yearly rejuvenating capability of the annual range. Indeed, most of California’s big game and upland wildlife populations rely on the annual forage crop for significant portions of their diet. The titure of these valuable resources and capital investments need not be wholly entrusted to the whims of nature’s erratically patterned and oftentimes meager rainfall. The purpose of this study was to show how prudent and knowledgeable seasonal management of the mulch layer in conjunction with an understanding of site differences over California’s 6 million hectares of annual range can significantly influence the subsequent year’s forage production. The 6 million hectares of grassland in California (Barbour and Major 1977) extend nearly the entire length of the state. Stand composition and productivity vary considerably over the grassland. Munz and Keck ( 1949) divided the grassland into two plant communities, the Coastal Prairie and the Valley Grassland. Kiichler’s (1964) Fescue-Oatgrass and California Steppe vegetation types as updated into Coastal Prairie-Scrub Mosaic and Valley grassland (Ktichler 1977) retain the basic divisions of Munz and Keck. Detailed descriptions of species composition within the annual grassland remain scarce. Burcham (1975) reported abundance of annual grassland species from 38 plots in the Sierra foothills from Sacramento to Madera County and in the Coast ranges from Santa Clara to Monterey County. Species of Bromus and Erodium exhibited a high degree of constancy. Soft chess (Bromus mollis) formed a significant amount of the vegetative cover on 30 plots and was absent on only one plot. Ripgut brome (Bromus rigidus), although a significant percentage of the cover on only one plot, occurred on 2 1 plots. Erodium species were present on 3 1 plots, forming a significant portion of the cover on 28 plots. Janes (1969) sampled species composition on the western side of the Central Valley and in the North Coast ranges along an average annual precipitation gradient ranging from 12.5 to 200 cm and also showed soft chess to be the most widely distributed annual grassland species. Janes encountered 124 species at his 20 sample locations, yet only seven species occurred at more than four locations. Soft chess occurred at 15 sites, ripgut brome at 11 sites, and broad-leaved filaree (Erodium botrys) at 7 sites. Janes’ study in the spring of 1969 remains the only comprehensive survey of geographical variation in the annual grassland. Understanding site variation in annual grassland is necessary for effective management. Composition and productivity of annual ranges vary remarkably between and within years. JOURNALOF RANGE MANAGEMENT33(1), January 1980 Talbot et al. ( 1939) and Bentley and Talbot ( 195 1) provided several years of data on yearly variations in composition and production of annual grassland at the U.S. Forest Serviceoperated San Joaquin Experimental Range (50 km north of Fresno), a site with continuous documentation since the early 1930’s (Duncan and Woodmansee 1975). Composition and productivity at the University of California’s Hopland Field Station has been monitored since the early 1950’s with documentation of variations within the year (Heady 1958; Bartolome 1978; Savelle 1977) and yearly long-term production and composition (Murphy 1970; Pitt and Heady 1978) in relation to weather patterns. Consideration of sites apart from the San Joaquin Range and Hopland are few, including Evans et al. ( 1975) at the University of California’s Sierra Foothill Range 100 km north of Sacramento, Batzli and Pitelka (1970), and Raliff and Heady (1962) near San Francisco Bay in Alameda County. The impacts of grazing by livestock on annual grassland have not often been directly investigated. Elliot and Wehausen (1974) examined response of coastal prairie species to grazing on Point Reyes. A long-term grazing study at the San Joaquin Experimental Range enabled the development of management guidelines for annual range (Bentley and Talbot 195 1). Hedrick (1948) also looked at the effects of grazing intensity on annual range. Most of the response of annual gr_assland species to grazing has been inferred from experiments simulating grazing through the removal of varying amounts of plant residue or mulch (Talbot et al. 1939; Heady 1956). Experiments at Hopland showed a response of the grassland to mulch manipulation similar to that which would be expected under livestock grazing (Heady 196 1). These results were extended to a discussion of economic factors (Hooper and Heady 1970) involved in intensity of mulch removal on annual grassland. Mechanisms explaining the effects of mulch have not been demonstrated, but several factors probably are important. Soil organic matter and bulk density are both improved with increasing amounts of mulch, but only in the upper few inches of the soil (Heady 1966). Mulch protects the soil surface from erosion and provides a favorable environment for plant growth. 70 o Fresno \ ’ 200km I Fig. 1. Mup oj Culijorniu showing location oj nine study plots Several authors have ascribed the differences in botanical composition due to manipulation of mulch to the effect on plant establishment (Bartolome 1978; Evans and Young 1970; Tinnin and Muller 197 1).

Opal phytoliths as evidence for displacement of native Californian grassland

Abstract Opal phytoliths are produced by plants and persist in soils as microfossils with taxonomically distinct morphology. We found phytoliths produced by the original native perennial species in soil under an annual grassland, thus providing the first direct evidence that the Californian grassland was formerly dominated by panicoid opal-producing grass. The most common panicoid-type opals were probably produced by Stipa pulchra. Frequencies of opal phytoliths from native grasses were greater at 10 cm depth than at the soil surface beneath the annual grassland. Comparison of opal phytolith frequencies from 10 cm deep at the annual site and an adjacent relict perennial grassland site suggested that the density of panicoid opal-producing native grasses on the annual site was once similar to the relict grassland.

Stable carbon isotope composition of Poaceae pollen and its potential in paleovegetational reconstructions

Stable carbon isotope differences between ecologically distinct groups of Poaceae (C3 and C4 photosynthetic groups) provide a means of isotopically subdividing grass pollen in paleovegetation studies. We examined the isotopic composition of bulk grass plant tissue, untreated pollen, and chemically treated pollen, from several C3 and C4 grass species. Based on our data, untreated pollen is isotopically similar to the host plant from which it is derived, although small, random differences between plants and pollen occur. Methods of pollen concentration involving carbon-bearing compounds can alter the isotopic composition of recovered pollen, and in some cases, make pollen from different grass types isotopically indistinguishable. We conclude that the isotopic composition of physically separated Poaceae pollen should be an important means of determining the proportion of C3C4 grasses as long a carbon-bearing chemicals are not used in sample preparation. The carbon isotope composition of pollen should provide a new means of determining paleoclimatic conditions in grassland environments and aid in identifying the origin of the C4 photosynthetic pathway in the geologic past.

Reciprocal transplants of herbaceous communities between Quercus agrifolia woodland and adjacent grassland

1) Oak woodlands in California intermingle with annual grasslands. They frequently have a grassy understorey, which has a distinct species composition and abundance unlike adjacent open communities. Soil blocks were reciprocally transplanted in autumn (at seed stage) between oak understorey and open grassland. Changes in species composition and species population densities were monitored over the following two years. 2 Species composition of adult plants in the reciprocally transplanted soil blocks resembled the original «donor» seed bank in the first spring after transplant, but in the second year it was not very different from the surrounding community in the «receptor» habitat

Ecological dynamics of Quercus dominated woodlands in California and southern Spain: a state-transition model

There are many similarities between Spanish and Californian Quercus woodlands and savanna. Both are located in Mediterranean climate zones, and are used predominantly for livestock grazing. The Californian overstory is dominated by one or a combination of five Quercus species and their hybrids: Quercus douglasii H.&A., Q. agrifolia Nee., Q. wislizenii A.DC., Q. lobata Nee., and Q. englemennii Greene (blue, coast live, interior live, valley, and Englemann oaks). In southern Spain and Portugal, Quercus woodland overstory is predominantly one or a combination of two Quercus species, Quercus ilex L. (holm oak) and Quercus suber L. (cork oak). The underlying natural and semi-natural ecological dynamics of the Quercus woodlands of Spain and California are different, and it follows that the management practices employed also differ. The greatest point of contrast between California and Spain is in the intensity and diversity of management goals and practices. A state-transition model for comparing the ecological dynamics of Quercus woodlands and savanna in California and southern Spain is developed and examined. The highly simplified model is an analytic tool of use in organizing research and developing management practices. States are reached and maintained in different ways in Spain and California, but their appearance and their function in each landscape are quite similar.

Efficacy of natural grassland buffers for removal of Cryptosporidium parvum in rangeland runoff

Our goal for this project was to estimate the retention efficiency of natural grassland buffers for Cryptosporidium parvum. Three sets of 16 plots (2.0 by 3.0 m) were established at 5, 20, and 35% slopes. Within each set of 16 plots, residual dry vegetation matter treatments of 225, 560, and 900 kg/ha were implemented, along with a noncut control averaging 4,500 kg/ ha. Buffer width treatments were implemented by placing cattle fecal material containing known loads of C. parvum 0.1, 1.1, or 2.1 m up-slope of the runoff collector. Grassland buffers of 1.1 and 2.1 m generated 3.2- to 8.8-log and 3.6- to 8.8-log retention of C. parvum, respectively, across the range of residual dry vegetation matter, land slope, rainfall, and runoff conditions examined during this project. Buffers with an increased percent land slope exhibited improved the retention efficiencies, whereas buffers experiencing larger maximum annual runoff events exhibited reduced retention efficiencies. Water-quality data from the 0.1-m-wide buffer plots (effectively no buffer) demonstrated that the majority of C. parvum oocysts (98 to 99.999%) were retained in the fecal matrix for the duration of the storm season, irrespective of the presence of a vegetated buffer. In conclusion, these results support the assertion that grassland buffers are an effective method for reducing animal agricultural inputs of waterborne C. parvum into drinking and irrigation water supplies.

Local Temporal and Spatial Structure

Community structure is defined by the organization of its parts. This paper describes plant species, but local temporal and spatial structure can refer to any definable organizational elements in the community. Since science is essentially a search for repeatable patterns, the identification of predictable structural elements must form the basis for explanations of function. For example, California’s grassland communities most commonly are described in terms of relative cover of the dominant species. Cover is traditionally emphasized because of a presumed functional relationship to ecological dominance. Any functional explanation depends on structural descriptions, a relationship infrequently acknowledged (Whittaker 1970), although resource partitioning is a fundamental tenet of ecological theory.

APPLICATION OF HERBIVORE OPTIMIZATION THEORY TO RANGELANDS OF THE WESTERN UNITED STATES

The central assumption for management of range condition-that plant response to selective grazing drives changes in plant community structure-is only weakly supported by evidence from semi-arid rangelands. Most of the vegetation changes attributed to selective grazing can instead be explained through proper interpretation of grazing intensity. Specialized livestock grazing systems, which assume that selective seasonal grazing controls ecosystem function, work poorly on semi-arid rangelands when compared to simpler grazing methods based on managing grazing intensity. Compensatory growth has been well linked to ecosystem processes in highly productive and intensively managed pastures, but not on semi-arid rangelands.