Precipitation Maps for Ranchers in the Rain Shadow of the Rockies
By Nicole Kaplan
“We shall never achieve harmony with the land, any more than we shall achieve absolute justice or liberty for people. In these higher aspirations, the important thing is not to achieve but to strive.” by Aldo Leopold
Background
We strive to adapt rangeland management to be in greater harmony with conditions in the Western Great Plains. Adaptive management incorporates various data sources to make decisions based on environmental as well as economic objectives. Adaptive management is an approach to increase the sustainability of livestock production as well as provision of ecosystem services for ecosystems and people who live in the rain shadow of the Rocky Mountains. The western Great Plains have evolved with grazing of native herbivores, such as the American Bison, and now is grazed by beef cattle. This area is also considered one of the most endangered grasslands of North America as habitat for species of grassland birds is degraded or fragmented by various land use practices (Davis et al. 2020). If ranchers can stay in business and manage their ranches for multiple objectives including profitability and conservation, then the cultural, ecological, and agricultural benefits of this area is more likely to be sustained. My Earth Data Analytics project focuses on one type of data that informs ranchers' decisions, and that is precipitation. The western Great Plains sit along the rain shadow of the Rocky Mountains, where dominant grasses are low growing and resistant to grazing. This is ideal for livestock production, which serves as the prominent agricultural practice on the landscape. The area is also prone to drought and high variability of rainfall. Most precipitation falls in spring and summer from convective storms which can rain unevenly across the landscape. Ranchers need data from across their ranches in order to manage for various local conditions (Raynor et al. 2020). Currently, state-wide datasets do not provide this level of detail. In my project, I use precipitation data collected at points within pastures to create an interpolated ranch level map within an experiment on the USDA Central Plains Experimental Range. Stakeholders involved in the Collaborative Adaptive Rangeland Management (CARM) experiment need to interpret precipitation data to make timely decisions. A description of the CARM experiment and CPER site follows.
*** **Rainstorm on the grassland showing local nature of rainfall on the grasslands, by Sean Hauser**
The Collaborative Adaptive Rangeland Management Experiment
Collaborative Adaptive Rangeland Management (CARM) on the USDA ARS Central Plains Experimental Range in northeastern Colorado sets up an experiment to examine how science can be conducted in a real-world manner (i.e., at ranch-level scales with manager involvement) to evaluate the effectiveness of adaptive grazing management for both production and conservation goals. In particular, we seek to examine how grazing management can be implemented in a manner that responds to current and changing rangeland conditions, incorporates active learning, and makes decisions based on quantitative, repeatable measurements collected at multiple spatial and temporal scales. For this project, precipitation has been collected across the CPER as a time-series field-based dataset. Precipitation timing and amounts drive vegetation growth, and vegetation must be available for cattle, as well as wild fauna, to thrive in a pasture.
CARM is an experiment with season long moderate grazing in ten pastures as the control, and a large herd rotated across ten treatment pastures. This real-world experimental design allows stakeholders to test various management decisions informed by data that are collected from satellites, sensors and field-crews and provided to them. A Stakeholder Group of 11 persons representing ranchers, public land managers, conservation organizations and nongovernmental organizations establishes triggers for when a large herd of cattle move from one pasture to the next, with their objectives for management being both beef production and conservation. Every week the stakeholders receive an update on precipitation amounts, as well as available forage to decide whether cattle should remain or be rotated. Presenting data in useful visualizations are essential to this process.
Cattle Drive for the Big CARM Herd on the Central Plains Experimental Range, Nunn, Colorado
The Central Plains Experimental Range Study Site
In the mid 1930's, many farms and ranches of the western Great Plains were abandoned due to drought, overgrazing and soil blown from plowed fields. The U.S. Forest Service requested that the Central Plains Experimental Range (CPER) be established to research improved management practices on fragile grasslands. The first research project was initiated in 1939 by the U.S. Forest Service. The Agricultural Act of 1953 reorganized the USDA and transferred administration of the CPER from the Forest Service to the Agricultural Research Service (ARS). Grazing studies, such as CARM, begun in the 1930's are still being conducted to evaluate the long-term impacts of livestock on rangeland resources. Many researchers from universities use the CPER as a study area, as well are large, collaborative research networks, such as the International Biome Program, Man and The Biosphere, Long-Term Ecological Research Network, National Ecological Observatory Network, and the Long-term Agroecosystem Research Network. Data produced by many of these network will be included in the Earth Data Analytics Capstone project, which will go beyond precipitation data and one year of values.
View from CPER towards the Rocky Mountains
Map of Study Area
The US Drought Monitor tells us if this region is in a drought:
The CPER sits within a region referred to as the High Plains. The US Drought Monitor performs data analysis on a weekly basis to determine which areas are experiencing a drought or vulnerable due to drying conditions. This is the latest map from June 23, 2020.
Findings: Precipitation Map for CARM Pastures
In 2020, by June 23rd, there is concern a drought is coming to the CPER area. Annual precipitation received at CPER has been seventy-six percent of the long-term average for this point in the year. CARM stakeholders managing the movement of the big herd as to adapt to changing conditions have defined local drought conditions to be seventy-five percent of the average cumulative rainfall. But, precipitation received has not been even across all pastures. Our interpolated map shows the variations in rainfall amounts, with the north east part of the station having received more rain. Without additional precipitation, available forage in pastures will decrease as it is grazed. Stakeholders must decide where to move the CARM herd to find sufficient food for weight gain and not negatively effect conservation habitat.
How can precipitation maps affect adaptive management decisions?
This timeline shows how the large herd, referred to as the CARM herd, is rotated through different pastures on the CPER. Rotation decisions are made by the 11 person stakeholder group (i.e. ranchers, conservationists, NGOs, land managers). Approximately 20 steers in 10 control pastures remain for the season at a moderate stocking rate, while the large herd of over 200 head are subject to adaptive grazing management are rotated over 10 pastures.
Stakeholders plan to keep the CARM herd in a pasture for twenty-one days, but decreasing forage could be a trigger for herd rotation. As the range has dried, the CARM herd has needed to be rotated more often. This timeline displays the date of movement as well as the pastures receiving the herd to help stakeholder keep track of herd movement.
On June 16th,the CARM herd is located in the pastures known as crossroads and south. These pastures are also targeted for McCown's longspur habitat, which require a shorter stature of vegetation. This could be a benefit to the birds, but these pastures are some of the driest locations on the CPER according to our precipitation map and the steers must be able to main their weight gain. So CARM stakeholders now must monitor the forage availability and decide if they should move the herd to pastures that have received more rain or stay with their planned rotation, which takes into account how far the herd needs to be moved, as well as using grazing to maintain habitat for birds.
On June 26th, the CPER experienced a deluge with some pastures receiving over 2 inches of rain, which provided some relief to stakeholders' concerns over drought. However the big herd moved to the next set of pastures on July 1 (see timeline of grazing rotations above) and if they continue grazing through pastures faster than planned this project may need to get adaptive in planning the rest of the season.
How we created a precipitation map for ranchers
My data source was data directly downloaded from a series of precipitation gauges across the CPER and processed for preliminary quality assurance on our ARS USDA server. A rolling total for the last three week of cumulative precipitation is calculated, to illustrate where potential growth and regrowth of vegetation may occur. Additional data types included vector data for pasture boundaries and the rotation schedule for the CARM herd, which comes from communication with our land managers. I used Inverse Distance Weighing (IDW) as an interpolation method to expand the visualization range of the point-based precipitation data to cover the entire Central Plains Experimental Range (Brodersen 2017). IDW computes a score of points in between locations of precipitation gauges, based on the scores of a number of near neighboring values. Scores or values are weighted by the inverse of their distances, but do not take into account any statistical relationships, such as auto-correlation. Functions were used to perform IDW interpolation, as well as classification of precipitation amounts.
The future of ranching: precision rangeland management driven by data
Merging precision environmental data (e.g. precipitation, soil water, remotely-sensed vegetation data), precision livestock management (e.g. GPS collars and instantaneous forage intakes, grazing behaviors), and population trends for conservation species of interest are part of the current works of scientists and staff supporting CARM to enhance decision making regarding provision of multiple ecosystem goods and services for sustainable rangelands. If maps like this could contribute to decision tools that stakeholders and researchers co-design, test, and refine with scientific programming then perhaps robust apps could be created for ranchers for use at their local ranch. These are innovative and collaborative advances to address one of USDA's strategic goals to “Strengthen the Stewardship of Private Lands Through Technology and Research”, which I see as a means towards striving to live in harmony with the land.
For scientists at the CPER, scientific programming provides scripts for automating workflows of data from the field to the stakeholders, allowing us to work smarter, not and harder, and spend more time on science. For stakeholders managing rangelands in the western Great Plains, it provides information and new approaches that can support adaptive management under varying environmental conditions in the western Great Plains.
We are looking forward to putting data from additional sensors through this workflow, sharing our approaches with sister sites in the USDA Long-term Agroecosystem Research network, as well as neighboring ranchers interested in precision rangeland management. Lastly, we can continue to explore different kinds of data interpolation, including kriging to improve our maps.
Link to my GitHub Repo for my scientific programming approaches to mapping precipitation.
Horned Lark in Landscape of the Central Plains Experimental Range, Nunn, Colorado, by Nicole Kaplan
