Our Science "Vision" Today

Twitter icon
Facebook icon
YouTube icon
RSS icon
Printer-friendly versionPrinter-friendly versionSend by emailSend by email

The Range Problem today calls not only for different experiments, but also for a reformulation of the broader framework of the discipline away from its modernist roots—productivist, reductionist, and mechanistic. The “postmodern” version of range science is the application of the scientific method at scales spatially and temporally relevant to management. We are trying to learn how to apply science to local management settings, with all their complexity, yet retain the scientific method of a post modern science generating data and analyses to support land management, and lands of all types.

First, we envision a science at the scale of landscapes, with all of their hitherto unidentified and uncontrolled variations. Spatially, this means embedding smaller units—from individual plants to plots, pastures and ranches—within the surrounding landscape and recognizing the interconnections among processes operating at different scales. Hierarchy theory can provide an initial framework for organizing observations and formulating hypotheses, but simple notions of “top-down constraints” or “bottom-up mechanisms” should be viewed as provisional and heuristic devices, not established facts.

Second, temporal scales should be treated in an analogous manner: processes operating at different time scales interact, sometimes resulting in abrupt changes in system attributes (thresholds) that endure as historic legacies, strongly determining the range of possible subsequent changes and management opportunities. Above all, the science we envision needs to be spatially and temporally explicit about the processes and interactions that we observe and seek to understand.

Third, we envision a scientific method that incorporates tests of hypotheses in "retrospective" experimental designs using landscape treatments of the past as an affordable means to work at scales relevant to management. In addition to biophysical events with historic legacies (e.g., severe droughts, unusually wet summers, or hard frosts), we must capture information about the human activities and management practices implemented on rangelands over the past century (or in some cases even longer). For example, if the dates and locations of construction of fences, wells and water developments, erosion control measures, or vegetation manipulations (e.g., bulldozing, chemical applications, or seeding) can be accurately ascertained, such practices can then be analyzed as experiments whose effects can be evaluated using both newly-gathered and historical records such as aerial photographs, monitoring data, and weather records. Such “experiments” will usually lack strict controls. However, if these landscapes can be delineated in an ecological fashion, possibly through ecological site boundaries, they can provide the landscape-scale contextualization described above. In this manner, past landscape experimentation can permit tentative comparison of management practice effects and effectiveness across time and space.   

Fourth, scientists must engage land management professionals and ranchers, on both public and private lands, as collaborators and sources of information and knowledge, recognizing their long-term experience in specific sites and the potential value of that experience for improving scientific understanding. Existing datasets can yield patterns at large scales, but such analysis cannot reveal the mechanisms that produced those patterns. Long-time residents and managers often possess site-specific documentation such as rainfall records, memories of when changes occurred, and knowledge of any unique or unusual events that preceded or accompanied changes (e.g., flash floods). They also provide insight into the social variables that acted as causes or preconditions for management interventions, such as periods of prosperity that allowed ranchers to make large investments in fences, waters, or vegetation manipulations. Such local knowledge can thus help to identify mechanisms of change to explain patterns discerned (or corroborated) from other sources. The resulting integration of diverse data, knowledge sources, and methods of investigation could be described as a science of consilience of facts and anecdotes across a landscape, thoroughly interdisciplinary in nature.

Fifth, we envision a science that is post-productivist and non-reductionist; a science that recognizes a diversity of social objectives, including both traditional commodities and (as yet) non-market “ecosystem services.” Scientists must weigh and prioritize among objectives in open dialogue with interested parties, and seek to understand and communicate the complexity of the biophysical processes that interact to produce valued goods and services. Here the scientific emphasis is on analytical services that exploit the rapidly increasing layers of available data rather than on science-based technologies narrowly tailored to discrete goods.

Sixth and finally, we envision a science that is public in multiple senses of the word: A science whose practices and data are transparent and accessible as broadly as possible; that serves public needs and interests and is receptive to public participation; that is applicable as one of many inputs to policy; and that is communicated in ways that enable it to contribute to those policies.

Implications

Given the above described retrospective philosophy and emerging tactics for range science after the past century of effort, we see a much different framework for the research questions that need to be addressed to move us towards this vision and beyond what Wooton and others envisioned over a Century ago (Table 1). There is an important implication within this framework. This implication is that conducting science in this way is considerably different from how we have conducted range science in the past. Not only are we arguing for science at larger spatial and longer temporal scales than we have addressed in the past, but a science conducted without the tidy experimental designs of our agronomic, highly-controlled, reductionist past. The science we are arguing for, and that the heterogeneities of these landscapes demands, is much less controlled, much more variable, and yet much more applied than practiced over the previous century. It is a science directed towards specific landscapes, informed by local knowledge and the management "experiments" that have been conducted, successfully or not, in the past and the present. It is a science rooted firmly in management as a hypothesis, i.e., the scientific method is used to test the effects of management practices as they are applied, or have been applied, in the field. In this fashion, the practices of management become the basis of scientific hypotheses, science provides the framework for testing those hypotheses, and the knowledge that results can readily feed back into management.

 

Table 1.  Contrasting challenges for range science before and after a century of range research.

Modern, 20th Century Range Science Post Modern, 21st Century Range Science
Identification of Cardinal Principles of management Contextualization of general principles to specific landscapes
Determination of livestock carrying capacities for discrete parcels of land Identification and integration of values of multiple goods and services across entire landscapes
Conservation and recovery of herbaceous forage species Management strategies for a world with less grass
Linear stages of plant communities in various but stable climates Adaptive strategies for non-linear change, extreme events and changing climates
Practices for controlling invasive species Strategies for adaptation to non-native species
Impacts of biophysical drivers Impacts of socioeconomic drivers interacting with biophysical drivers
Linear transfer of scientific knowledge to users Scientists informed by local knowledge
Stored reductionist data for controlled analyses Accessible data for open analyses
Minimal spatial dimensions to experimentation Landscape scale dimensions to experimentation
Management not integrated into experimental designs Management as an integral part of hypotheses