Development of a spatially explicit ruminant landscape use model (RLUM)

Abstract

This paper describes the development of RLUM, an integrated spatially explicit, prototypic, ruminant landscape use model. RLUM integrates several component models: weather, plant growth, ruminate nutrition, diet selection and physio-behavior with a geographic information system (GIS) based animal movement and foraging model. In addition to housing the movement and foraging simulation model, the GIS enables us to explicitly represent landscape features. RLUM simulates landscape use for an entire herd (95 head) of cattle. Historical weather files were used as input into the plant growth and physiobehavior models. The physio-behavior model determines an animal's current physiological status in terms of hydric, thermal, and hunger balance equations. Information on the animal's physiological status is used to determine its behavior, which in turn determines where the animal will move on the simulated landscape. Several aspects of model performance were evaluated using field data collected as part of this study as well as published literature. Simulated landscape use patterns show a tight pattern concentrated around water sources whereas observed landscape This paper describes the development of RLUM, an integrated spatially explicit, prototypic, ruminant landscape use model. RLUM integrates several component models: weather, plant growth, ruminate nutrition, diet selection and physio-behavior with a geographic information system (GIS) based animal movement and foraging model. In addition to housing the movement and foraging simulation model, the GIS enables us to explicitly represent landscape features. RLUM simulates landscape use for an entire herd (95 head) of cattle. Historical weather files were used as input into the plant growth and physiobehavior models. The physio-behavior model determines an animal's current physiological status in terms of hydric, thermal, and hunger balance equations. Information on the animal's physiological status is used to determine its behavior, which in turn determines where the animal will move on the simulated landscape. Several aspects of model performance were evaluated using field data collected as part of this study as well as published literature. Simulated landscape use patterns show a tight pattern concentrated around water sources whereas observed landscape use patterns were more diffuse. Differences in the observed vs. simulated landscape use patterns can be explained by the fact that an entire herd and not individuals or subherds were modeled from a single water point. Also, the initial starting conditions (i.e. forage availability and distribution) resemble that of a pasture that was deferred from grazing, yet in reality the study pastures were continuously stocked. This study has shown that simulation models and geographic information systems can be integrated to simulate landscape use by cattle. In addition, RLUM has established a framework for future development of multi-species, multi-scale landscape use models.