Understanding Implementation of Adaptive Management

Adaptive Management of Rangelands Using Precision Agricultural Technology
New tools and technology are increasingly available to agricultural producers to assist with increasing the efficiency and productivity of their operations. One such tool now available to ranchers is virtual fencing, a technology designed to allow ranchers to manage, move, and confine livestock herds without the need for physical fences and with reduced need for human labor. However, the technology is new and there are still many questions about its effectiveness, affordability, and impacts on animal welfare. Our lab is investigating these questions: how well does virtual fencing work as a range management tool and can it be used to implement adaptive management, what are its impacts on animal welfare or the short and long term, and what is the economic viability of virtual fencing for ranching operations and what is the likelihood of adoption of virtual fencing by ranchers? This work is funded by the National Institute for Food and Agriculture Interdisciplinary Explorations in Animal Systems (IDEAS) grant to the University of Arizona.

Responding to Emerging Threats to Coupled Natural Human Systems
Exogenous threats, including climate change, invasive species, and exotic diseases threaten the stability and resiliency of coupled natural and human systems. These challenges are particularly vexing for resource managers because they have incomplete knowledge of existing systems will respond to these changes and because existing policies can limit adaptive responses. Invasive plants are one example of this problem: the invasive species enters a new area where land managers have no knowledge of how to mitigate its spread, no understanding of its potential impacts, and no existing capacity to coordinate to prevent impacts. I am working with a team of ecologists and social scientists to understand how managers respond to emerging threats, use agent-based modeling to predict outcomes of these responses, and co-produce adaptive responses with managers based on the results of the predictive modeling. This work is funded by an NSF Coupled Natural and Human Systems (CNH) grant to the University of Arizona.

Adaptive Management on Public Lands
Public lands have a wide range of uses, from grazing and timber extraction, to wildlife habitat, to recreation. To improve its ability to manage these overlapping and sometimes competing uses, the US Forest Service has embraced adaptive management. This work seeks to understand the impacts of the Forest Service’s use of adaptive management to administer grazing allotments in Arizona and New Mexico. Research methods include interviews with ranchers and Forest Service personnel, a large-n survey of ranchers throughout the region, and analysis of thousands of annual allotment management documents to identify changes in allotment administration. This work is ongoing, but early results show that effective implementation of adaptive management is dependent on high levels of communication and trust between permittees and Forest Service personnel. This work was supported by a National Institute for Food and Agriculture grant to the University of Arizona.

Making Sense of the National Environmental Policy Act with Big Data
The National Environmental Policy Act, aka NEPA, is the cornerstone of US environmental law. It requires the evaluation of the impacts of any significant federal project on the environment and people. Over 50 years, NEPA has caused the creation of tens of thousands of environmental impact statements. However, there has been little evaluation of the law’s effectiveness at achieving its intended purpose – to reduce environmental impacts from development. To address this question, I am working with a team of data scientists and social scientists to develop a knowledge platform for evaluation of NEPA’s effectiveness. This platform will include documentation from tens of thousands of NEPA document and use tools such as machine learning and natural language processing to extract data. This work is funded by an NSF Resource Implementations in Data Intensive Research (RIDIR) grant to the University of Arizona.