LIVESTOCK RANCHING, RANGELANDS, AND RESILIENCE: ENSURING ADAPTIVE CAPACITY IN AN INCREASINGLY VARIABLE CLIMATE

• Sponsoring Institution: National Institute of Food and Agriculture
• Project Status: NEW
• Funding Source: AFRI COMPETITIVE GRANT
• Reporting Frequency: Annual
• Accession No.: 1015486
• Grant No.: 2018-68002-27923
• Project No.: COLW-2017-07324
• Proposal No.: 2017-07324
• Program Code: A3171
• Project Start Date: May 1, 2018
• Project End Date: Apr 30, 2022
• Grant Year: 2018

Project Director
Suding, K.

Recipient Organization
UNIVERSITY OF COLORADO
(N/A)
BOULDER,CO 80309

Performing Department
INSTAAR

Non Technical Summary
Droughts are one of the most devastating natural hazards faced by the United States today. Droughts more severe than those of the 1930s and 1950s are likely to occur with greater frequency in the future. These increases in drought frequency and intensity will be accompanied by more intense storms and record wet periods: witness the recent mega-drought in California, broken by a year of unprecedented rainfall. The challenge is not just coping with drought, but ensuring sustainability in a future of both droughts and deluges.The need to plan and implement adaptation strategies is perhaps nowhere more urgent than in the rangeland sector. Rangeland production systems are globally important and highly vulnerable to current and future climate conditions.Managed grazing covers more than 25% of the global land surface, a larger geographic extent than any other form of land use. In the US, 300 million ha of public and private lands are rangelands. Rangelands provide society with goods (e.g., livestock production/food), services (e.g., wildlife habitat, conservation, water quality), and contribute to the livelihood of millions of humans. However, rangelands are typically located in semi-arid and arid regions characterized by low plant productivity, high precipitation variability, and frequent drought. Rangeland managers often have limited financial and social capital, and in many cases are removed from policy makers and governing institutions.Recently the Sustainable Rangeland Roundtable convened university and agency researchers, public and private land managers and producers, and non-governmental organizations to jointly chart a research agenda of usable science for rangeland sustainability. Out of 142 challenges they identified, "understanding and managing for variability" ranked first. Here, we continue this coproduction process - involving intended end users throughout our research enterprise - to develop a framework to assess the sensitivity of rangeland production systemsto climate variability and identify strategies to improve the adaptive capacity in these systems.Our premise is that strategies to implement climate adaptation will be far more effective if they are tailored to local diversity in exposures, sensitivities and adaptation opportunities faced by ranchers and land managers.Some rangelands may simply experience less climatic variability than others, even in the future. Some rangelands may support forage plants particularly well-adapted to variability, while in others forage production may be sensitive to variability but livestock operations have effective strategies for coping with variability. This project will considerdifferences in vulnerability as key considerations in building adaptation strategies.We use a comparative framework that spans five rangeland regions within the Western US. Our approach combines climate modeling and assessment with forage and livestock production models in a co-development framework to build scenarios and assess feasible, effective adaptation strategies.This iterative process combines research and extension to support a co-development of ideas, capitalizing on diverse adaptation strategies across western rangelands. Our outcome will be a strategic framework that reflects the diversity of exposures, sensitivities and capacity to adapt across the Western US, and will result in a handbook, a white paper and interactive tool co-designed by the science and user communities involved.

Animal Health Component

0%

Research Effort Categories

Basic

70%

Applied

30%

Developmental

(N/A)

Classification

Knowledge Area (KA)	Subject of Investigation (SOI)	Field of Science (FOS)	Percent
102	0799	1070	40%
132	0799	2070	10%
801	0799	3080	30%
121	0799	3100	20%

Knowledge Area
102 - Soil, Plant, Water, Nutrient Relationships; 132 - Weather and Climate; 801 - Individual and Family Resource Management; 121 - Management of Range Resources;

Subject Of Investigation
0799 - Rangelands and grasslands, general;

Field Of Science
2070 - Meteorology and climatology; 1070 - Ecology; 3080 - Sociology; 3100 - Management;

Keywords

adaptive capacity

climate adaptation

coproduction

livestock production

rancher

rangeland

Goals / Objectives
More frequent drought and intense storms require adaptive management practices that enhance resilience to climate variability and stabilize agroecosystem production. Strategies to implement climate adaptation will be far more effective if they are tailored to local diversity in exposures, sensitivities and adaptation opportunities faced by ranchers and land managers.We have three specific objectives:1) EXPOSURE. Determine climate exposure in different Western US rangeland regions and project their exposure in the future.2) SENSITIVITY. Determine forage and livestock production sensitivities to increased climate variability, considering non-linearity and tipping points.3) ADPATIVE CAPACITY. Evaluate effectiveness of different adaptive strategies given regional exposures and sensitivities. We use a place-based coproduction approach, embedding our extension efforts throughout the project framework, to understand on-going adaptations to recent climate variability by the innovators and early adopters and then to explore how those early adaptations may need modification under future climate scenarios.

Project Methods
We focus on five rangeland regions in the Western US: California annual grasslands, cold deserts, northern mixed prairie, shortgrass steppe, and hot deserts.These broad vegetation types represent considerable diversity in precipitation seasonality and variability, temperature, plant functional type composition, land ownership, and production systems. Thisdiversity provides us with opportunities to test hypotheses about the factors selecting natural and social strategies for coping with variability, and to turn this understanding into effective programs to increase adaptive capacity. Focal sites within each region will anchor our work to long-term monitoring datasets, field expertise, and local extension specialists. Our comparative approach capitalizes on geographic differences in current and future climate conditions; we accompany this regional comparison with attention to within-region variability driven by differences in soils, vegetation, disturbances, and local actors.Objective 1: climate exposure1.1 Assess Current Climate Exposure. To evaluate current climate exposure across the focal rangeland regions, we will use climate records from each focal site to characterize trends in mean climate conditions, and interannual climate variability.1.2 Assess Climate Change Impacts. To evaluate future changes in climate, we will use downscaled climate projections. Because a goal is to relate this exposure to human decision-making, we will consider both a short timescale (20 years, within human memory), and a longer timescale (late 21st C, more typical of climate projections).1.3 Quantify Current and Future Soil Moisture Dynamics. To characterize soil water availability patterns, we will use an ecosystem water balance model (SOILWAT), parameterized with soil properties and vegetation composition data for each site and plot1.4 Using Exposure Results in the Co-Development Process. We will incorporate climate exposure into the co-development process in two ways. For the first workshop with producers and managers, we will develop information about exposure in a focal area over the previous 20 years. These data will inform initial interactions with producers and managers to discover what the innovators and early adopters have already done. For the second workshop, we will develop information about future climate scenarios that describe a gradient in business-as-usual (no change), increased variability representative of the focal region in 20 years, increased variability in the region with greatest exposure, and an extreme end of the 21st C scenario.Objective 2: sensitivity2.1 Determine Sensitivity in Forage Production. Using the soil water modeling results generated under Obj. 1, we will regress forage production on growing season soil water availability. Data on plant production, the response variable, will come from two sources. First, we will use direct, field-based estimates of aboveground net primary production collected at sites in each of our focal regions. Our second data source will be remotely-sensed gross and net primary production data derived from Landsat imagery.2.2 Project Future Forage Sensitivities. The production response functions represent a direct measure of climate sensitivity. To integrate climate sensitivity and exposure, we will project the effect of expected future changes in climate variability on the means and variances of forage production.2.3 Determine Sensitivity in Livestock Production. We will use model-based estimates of livestock production, rather than empirical data sets, driven directly by input data on forage quantity and quality, control for variability in management by focusing on "potential livestock production."2.4 Model Forage Quality. Our starting point for modeling the response of forage quality to precipitation is the data set of Craine et al. (2017), available on Data Dryad. The data set consists of more than 36,000 observations of forage quality (crude protein and digestible organic matter concentration) spanning a 22 year period and distributed across the western US.2.5 Model Potential Livestock Production. We will use a nutritional balance model (NUTBAL) based on the NRC system for determining nutritional requirements and intake rates and further modified by the Angerer lab.2.6 Using Sensitivity Results in Co-development Process. We will incorporate sensitivity into the co-development process (see section 4.3.2 for more details) in two ways. For the first workshop with producers and managers, we will develop information about production and livestock sensitivity in each focal area over the previous 20 years (e.g., from 4.2.1 and 4.2.3). These data will inform the initial interactions with producers and managers about the range of sensitivities that they can expect in their system. For the second workshop, we will combine the future climate scenarios with future sensitivities to allow exploration of possible futures. We will also assess the last component of sensitivity - that of rancher decision making - as part of the second workshops with individual managers.Objective 3: adaptive capacity3.1 Using Production Response Functions to Assess Forage Vegetation Adaptation. Our first step is to consider the variability in sensitivity functions obtained in the forage modeling objective (see section 4.2.2, above). These patterns will be to assess how sensitivity might changes due to species invasions, woody encroachment, disturbances and some types of grazing managements.3.2 Field Campaign to Characterize Vegetation Strategies. We will identify a suite of 20 sites within each focal region that show differences in forage production sensitivities along the lines described above. At each site, we will broadly characterize dominant functional groups, collect leaf tissue samples to assess water use efficiencies (13C), and soil samples to assess seed dormancy-related mechanisms related to bet-hedging. We will also collect seed of 3-4 of the dominant plant species to confirm the patterns measured in the field in greenhouse assessments.3.3 Scenario Analysis to Understand Livestock Producer Adaptation. A co-development process is an excellent vehicle to facilitate the interactions needed to advance the development and use adaptation strategies in the Western ranching livestock sector. Our co-development process will take place over the course of two cumulative workshops in each of the focal research sites.

Progress 05/01/18 to 04/30/19

Outputs
Target Audience:Our objectives stem from needs jointly identified by a diverse target audience: university and agency researchers, public and private land managers and producers, and non-governmental organizations. We expect our work to develop strategic guidance for ranchers and land managers so that they can identify effective adaptation strategies to climate variability given their geographic, ecological, and social situation. Extension work in each of five rangeland regions in the Western US (California annual grasslands, cold deserts, northern mixed prairie, shortgrass steppe, and hot deserts) will allow for the development of targeted, local guidance tailored to specific climates, vegetation, and social needs for ranchers. For policy makers and nongovernmental organizations, we will produce a white paper on the integration of regional and local exposure to guide diverse adaptive strategies in the western US. For university and agency researchers, we will produce peer-reviewed publications describing the outcomes of each of our objectives as well as their integration. Changes/Problems:There are no major changes to report. What opportunities for training and professional development has the project provided?This project has trained three postdoctoral researchers. Robert Shriver(USGS) has been active in the EXPOSURE group objectives, actively participates our bi-weekly meetings and our group meeting November 2018. He is primarily mentored by John Bradford. Andrew Felton (Utah State) has been active in the SENSITIVITY group objectives, actively participates our bi-weekly meetings and our group meeting November 2018.He is primarily mentored by Peter Adler. Christine Greene (UnivArizona)has been active in the ADAPTIVE CAPCITY group objectives, actively participates our bi-weekly meetings and our group meeting November 2018. She is primarily mentored by Mitch McClaran and Dan Ferguson. How have the results been disseminated to communities of interest?We are just completing the first year of this project, and so results have not been disseminated to communities of interest. We are planning our initial stakeholder meetings to begin May 2019. What do you plan to do during the next reporting period to accomplish the goals?1. EXPOSURE We will incorporate climate exposure into the co-development process in two ways. For the first workshop with producers and managers, we will develop information about exposure in a focal area over the previous 20 years. These data will inform initial interactions with producers and managers to discover what the innovators and early adopters have already done. For the second workshop, we will develop information about future climate scenarios that describe a gradient in business-as-usual (no change), increased variability representative of the focal region in 20 years, increased variability in the region with greatest exposure, and an extreme end of the 21st C scenario. 2. SENSITIVTY We haveinitiated our collaboration with Dr. Jay Angerer (Texas A&M) to analyze the response of forage quality to climate variability. That is the first step in our analysis of the sensitivity of livestock production to climate variability, which is a major goal during the next reporting period. 3. ADAPTIVE CAPACITY We are in the process of planning the first workshop with producers and managers, we will develop information about exposure in a focal area over the previous 20 years. We will be travelling to each of the 8 focal areas for these meetings in the next few months.

Impacts
What was accomplished under these goals? 1. EXPOSURE. We applied an ecosystem water balance model (SOILWAT2) to estimate detailed historical and expected future patterns of soil moisture and plant water availability. We simulated these conditions on a ~7km grid from for three time periods: 1915 to 2015 (using gridded datasets based on weather observations), 2020-2059, and 2060-2099 (based on downscaled climate projections from general circulation models.) SOILWAT2 is a daily time-step, multiple soil layer, mechanistic model of ecosystem water balance that accounts for plot-specific interactions between climate, soil conditions, and vegetation, to estimate water pools and fluxes. Soil texture data was gathered from ISRIC WISE 30 arc second database (Batjes 2016). Potential plant biomass at each plot was estimated using algorithms within SOILWAT2 that account for the relationships between long-term climate patterns and vegetation structure (Bradford et al. 2014). Historical daily temperature and precipitation data was derived from the Livneh gridded dataset across the western US (Livneh et al 2013). Multi-model ensembles of future conditions were developed using 2 representative concentration pathways (RCPs 4.5 and 8.5) to represent uncertainty in future CO2 emissions and 11 different GCMs that were strategically selected to both perform well for the western United States and represent variability in the entire suite of available GCMs. We analyzed these results to quantify the exposure of western U.S. rangelands to changes in climate (figure E1) and ecological drought (figure E2), including changes already observed and those expected in coming decades (figure E3). We are evaluating a broad array of potential drought metrics, including precipitation: total water year precipitation; soil water availability: average available water (i.e. above -3MPa) in each season; simultaneous hot and dry conditions: water-year cumulative degree when soil water potential is <-3mpa; and growing season timing: days on which 50% and 90% of water year transpiration is reached. These results will provide ecologically relevant metrics of drought for evaluating rangeland climate sensitivity, and will provide useful perspective for stakeholders in our outreach efforts. 2. SENSITIVTY The majority of progress to date has concerned processing the remotely sensed annual net primary productivity (rangeland plant growth) data in preparation for climate sensitivity analyses. We processed these data to assure our sensitivity analyses reflect responses of rangeland ecosystems, and not irrigated agricultural regions, based on satellite imagery. Processing occurred within Google Earth Engine and R software. We now have produced a 30-year (1986-2015) annual net primary productivity dataset of the western United States rangeland sector (at 7 km spatial scales). We have combined these data climate and soil moisture covariates produced by the Exposure group to produce a spatially referenced dataset ready for our analysis of forage production sensitivity. This dataset and the associated R computer code will be made publicly available to the research community. Currently, the data are stored on Google Drive and the computer code is stored on local hard drives and synced to Github, consistent with our data management plan. Our first analyses have focused on understanding how sensitivity of forage production to annual precipitation varies across distinct rangeland regions (i.e., vegetation types) of the western US. When we map the sensitivity of production to precipitation in individual locations (pixels), we see a general trend of decreasing sensitivity from east to west trend, with a notable exception in California annual grasslands. More generally, we are finding that locations that are more productive on average are less sensitive to interannual variation in precipitation, though the slope of this relationship varies markedly among vegetation types. Our next analyses will focus on how responses of production to our soil moisture covariates compare across vegetation types. This provide a more mechanistic understanding of how and why rangelands may differ in their sensitivity to future changes in the amount and timing of water availability expected under future climate changes. We have also worked with theoretical population models to test our intuition about how evolutionary strategies for coping with variability depend on both the magnitude and the predictability of environmental variation. The results mostly support the arguments we made in the proposal, with conservative bet-hedging strategies favored where variability is large in magnitude but unpredictable, and plasticity favored where cues provide some degree of predictability. However, we learned that these results depend on the form of density-dependence. We are now gathering empirical data sets to locate natural plant communities in this parameter space. 3. ADAPTIVE CAPACITY. The overarching goal for this objectiveis to inform the development of science that can increase rangeland adaptive capacity. To meet this goal, the HD team developed a strategy for engaging with ranchers, rangeland managers, cooperative extension agents and similar conveners to better understand how the social and biophysical systems on rangelands have been changing over the past 20 years as a benchmark for understanding recent adaptive capacity and how that may be manifest in the near future. During the reporting period, we conducted a literature review of the climate vulnerability of ranching across the Western United States as well as adaptive capacity, including both short-term and long-term drought adaptation strategies. We also worked with the research team to identify 8 focal landscapes for a comparative analysis of local knowledge and rangeland adaptive capacity. The identified focal landscapes include Central Arizona (Gila County), Southwest New Mexico (Luna, Sierra, Grant, and Hidalgo Counties), Northeast Colorado (Weld, Logan, and Morgan Counties), Central Nebraska (Cherry and Grant Counties), Southern Utah (Garfield and Kane Counties), Northwest Nevada (Humboldt County), and Northeast Utah (Rich County). These landscapes were selected as they represent a range of climates, forage ecology, and ranching on both public and private land. We developed the research protocols for the focus group discussions and surveys and applied for Institutional Review Board (IRB) approval of human subjects research. The protocol for the upcoming focus group discussions was developed to identify: 1. local knowledge of the rangeland context, 2. activity and decision calendars, including how climate variability changes activities and decisions in ranching and rangeland management, and 3. ranking of threats, both environmental and social, to ranching and rangeland management.

Publications