UC Davis Department of Plant Sciences

Ecosystem Service-based State and Transition Models

Keywords – oak woodland, oak savanna, annual grassland, Sierra Nevada, brush management, ecological site description, conservation practice effectiveness assessment, grazing management

ParticipantsLeslie Roche, D.J. Eastburn, James Chang, Toby O'Geen, Valerie Eviner

Any Comments or Questions?


Brush Management and California's Oak-Grassland Ecosystems
California's oak-grassland ecosystems cover approximately 6.4 million hectares, and produce 70% of the state's forage base, which helps support a $3 billion/year beef cattle industry. About 85% of California's surface drinking and irrigation water supply is generated or stored in these ecosystems. This is one of the most species-rich ecosystems in California, supporting over 300 vertebrate, 5000 invertebrate, and 2000 plant species.



    There has been large scale mechanical-chemical tree and shrub thinning within oak-grassland ecosystems for commercial fuel wood harvest, brush management and cattle forage production goals. Across the landscape, these management practices have increased carrying capacities for grazing cattle. These practices create site conditions that may deliver different levels of ecological services – when considered in the context of the state and transition concept. For example, a management unit could be transitioned, by management, from oak woodland to open grassland site and thus provide less carbon sequestration but greater cattle forage production. Forces such as climate change, fire, and invasive species can also transition a site from one state to another.

Given the oak-grassland’s economic importance and multiple ecosystem services this system supports, the challenge is to simultaneously manage for multiple goals, recognizing the value of all the services to society. State and transition succession models are one way of describing the various states and associated  ecosystem services that a particular site can achieve, the forces that can  transition a  site between states, and the role that management plays in the process. Current expert models are organized primarily on plant communities, rather than directly on ecosystem services such as carbon sequestration, water supply, and biological integrity. This project is about explicitly incorporating ecosystem services into these models to guide management and restoration.
Project Objectives
1) Identify ecologically unique sites and states based upon plant-soil dependent ecosystem services – as opposed to just upon vegetation properties.
2) Identify cost-effective and easily measurable proxies for dynamic soil and vegetation properties supporting key ecosystem services – allowing easy identification of ecologically distinct sites and states.
3) Quantify associations between these proxies/properties, ecosystem services and common integrated oak and grazing management scenarios – assess the potential for range management to trigger transitions.

Study Site
We are conducting this project at the UC Sierra Foothill Research and Extension Center, which is part of the Sierra Nevada Gravelly Loam Foothill ecological site. This site broadly represents about 2 million ha in CA. The study is a cross-sectional, longitudinal survey of vegetation and soil-based properties, proxies for these properties, and plant-soil dependent ecosystem services. The survey is being conducted across potentially different states that have been created by a gradient of integrated woody species management and resulting increases in cattle grazing/carrying capacity (Table 1).



Table 1. Mean cattle stocking rate from 1998 through 2008 for enrolled management units (pastures) within each management intensity category. AUY is animal unit year – ha/AUY is the number of hectares per animal unit (1000 lb mature cow) allocated for each year.


    Figure 1. Aerial photo of UC SFREC illustrating oak woodland, oak savanna, and open grassland states on the Sierra Nevada Gravelly Loam Foothill ecological site.

Expert STM Model for the Sierra Nevada Gravelly Loam Foothill Ecological Site
STMs are developed for individual ecological sites (soil-climate based land divisions). Ecological sites can support collections of alternative stable states, consisting of one or more plant community phases, which represent within-state variation. Community phase shifts within a state may be driven by natural events (e.g., multi-year drought) and/or management activities (e.g., oak thinning). States are separated by thresholds, which theoretically represent the points at which ecosystem structure and function have been sufficiently modified. That is, the system has been modified beyond the initial state’s ecological resilience - the system’s capacity to respond to disturbance while maintaining the same functions, structures, and identity, pushing the system to a different state. Biotic and abiotic variables that act as mechanisms of transition, contributing to the loss of stabilizing processes (i.e., negative feedbacks) and ecosystem resilience, are termed triggers.

Technical Definitions (From Briske et al. 2008)
State. A suite of plant community phases occurring on similar soils that interact with the environment to produce persistent functional and structural attributes associated with a characteristic range of variability.

Threshold. Conditions sufficient to modify ecosystem structure and function beyond the limits of ecological resilience, resulting in the formation of alternative states.

Restoration pathways. Re-establishment of prethreshold states following active restoration of negative feedback mechanisms necessary to maintain the resilience of these states.


Figure 2. Current plant-community based expert state and transition model for Sierra Nevada Gravelly Loam Foothill ecological site at UC SFREC  

Study Design and Sampling Strategy
This is a cross-sectional, longitudinal survey of multiple vegetation states existing across this site. States were created by intensive mechanical and chemical woody species/brush management activities during the period 1960 through 1975. Transition of a management unit (pasture) from oak woodland to one of the less wooded sites was followed by an increase in cattle stocking rate for that unit (Table 1, Figure 1).

Vegetation and soil properties, proxies for these properties, and plant-soil dependent ecosystem services listed in Table 2 are being sampled using a hierarchically nested (four levels) plot-based sampling design—a modified Whittaker plot: 1x1m (1 m²) subplots are nested within 10x10m (100 m²) plots, which are nested within a single 20x30m (600 m²) macro plot. Each modified Whittaker macro plot contains 6 plots and 12 subplots. This nested plot design and multiple plots per management unit will allow us to examine each project objective from multiple spatial scales (e.g., patch, management unit, ecological site). A total of 65 modified Whittaker plots were stratified across management units at SFREC representing the management intensity gradient illustrated in Table 1 and Figure 1. All services, properties, and proxies listed in Table 2 are measured on 15 of these Whittaker plots. Those 15 Whittaker plots contain a total of 25 intensely monitored soil profiles/subplots (2 to 3 intensely monitored soil profiles/subplots per modified Whittaker plot). At each of the 25 intensively monitored soil profiles/subplots, we have installed volumetric soil moisture probes, soil temperature probes, and anion/cation exchange resins to examine seasonal and annual patterns in these temporally dynamic soil properties. The remaining 50 Whittaker plots (i.e., those without intensely monitored soil profiles/subplots) are monitored for only a subset of the services, properties, and proxies listed in Table 2. Seasonal patterns of biomass accumulation and herbaceous plant community composition are measured via repeated sampling on all 65 modified Whittaker plots during the growing season.

Table 2. Ecosystem services, dynamic vegetation and soil properties, and potential rapid assessment proxies sampled during the study.  


Preliminary Results
Here we present preliminary ecosystem service-based results for a simplified state and transition model for this ecological site (Figure 3), focusing on oak woodland, oak savanna, annual grassland, and medusahead invaded annual grassland states. These results reflect values measured at the modified Whittaker macro plot scale (600 m²). Data collection and complete analysis are on-going for this project.

    Figure 3. A simplified expert state and transition model for the Sierra Nevada Gravelly Loam Foothill ecological site for the purpose of reporting preliminary results for the oak woodland, oak savanna, annual grassland, and medusahead invaded annual grassland plant community states.
Ecosystem Service – Nutrient Cycling; Property – Soil N and C
Retention and cycling of nutrients is an important service of this ecosystem. Total soil nitrogen and carbon levels reflect the relative retention of these key nutrients by the potential states on this site. Both N and C levels drop as a site transitions from woodland to savanna to annual grassland. There was no difference between annual grassland and annual grassland invaded by medusahead (Figure 4).
    Figure 4. Differences in total soil nitrogen (TN) and total soil carbon (TC) between states.

Ecosystem Service – Water Supply; Property – Infiltration; Proxy – Bulk Density
Water supply is a critical service this ecosystem provides to the people of California. This service is dependent upon soil hydrologic function, starting with soil surface infiltration. We are finding that infiltration drops as a site transitions from woodland to savanna to annual grassland. There was no difference between annual grassland and annual grassland invaded by medusahead (Figure 5).

We need easily measurable, cost effective, reliable proxies for key properties such as infiltration. Soil surface bulk density is a good proxy for infiltration. As bulk density increases we see a corresponding decrease in infiltration (Figure 6). However, we need relevant dynamic soil properties/proxies to make inferences about ecosystem service provisioning. While both infiltration and bulk density show dramatic response to changes in state on this site, these changes may not represent a functional difference in the ecosystem service of water supply. The frequency and intensity of storms in this region are such that a rainfall rate of 4 cm/hr occurs only once per 100 years. Given 'reduced' infiltration capacity of ~20 cm/hr for the annual grassland state - functional water supply differences may not exist between any of these states.



Figure 5. Differences in soil surface infiltration capacity between states measured by falling head infiltrometer.   Figure 6. Soil surface bulk density as a proxy for infiltration which is a determinant of the water supply ecosystem service.



Ecosystem Service – Biological Integrity;Property – Biodiversity; Proxy – H’
Biological integrity is the capacity of the community to support ecological processes, resist loss in capacity and recover. This service is approximated by herbaceous biodiversity (Shannon-Wiener Diversity Index H’, log scale). Herbaceous diversity declines as a site transitions from woodland to all subsequent states (Figure 7).

    Figure 7. Differences in herbaceous diversity between states measured by Shannon-Wiener Diversity Index (H’).  
  Ecosystem Service – Agricultural Productivity; Property – ANPP
This ecosystem provides 70% of the forage base for California’s beef cattle herds. Herbaceous forage production increases as a site transitions from wooded conditions (i.e., woodland and savanna states) to type-converted open grasslands (i.e. annual grassland and invasive annuals states)(Figure 8). Table 1 illustrates the differences in agricultural productivity provided by each of these states as measured by the carrying capacity (area required to feed a single cow for an entire year) of each state for this site.
Figure 8. Differences in forage production (herbaceous aboveground net primary productivity) between states.    


Management Implications
Expert state and transition models solely based on vegetation community likely will not reflect the functional differences/similarities in ecosystem services between potential states. In particular, the tradeoffs in services may not be fully realized as managers actively transition a site between potential states (i.e., via brush management, oak planting and restoration, weed control). STMs can be developed to reflect functional differences/similarities and will be valuable in organizing information for managers interested in managing for multiple outcomes. Woody species management and brush management is still actively practiced within this ecosystem. Across the landscape, management practices and supporting programs can have differential impacts on ecosystem services given the variation in regulations/ordinances among counties.


© 2011-2012 UC Davis | California Rangeland Watershed Laboratory | One Shields Ave | Davis, CA 95616 | Last update: November 18, 2015