In rangeland and savannah environments pastures are highly heterogeneous, with mixed plant species of different phenology, a wide range in biomass and the amount of exposed soil background. The spatial scale of many remotely-sensed images is too coarse to represent this heterogeneity. In tropical environments, the predominance of tall ��tussock�� grasses makes ground-based measurement of biomass difficult. We focus here on the remote sensing of pastures rather than other landscape features such as trees.Within herbivore grazing systems, independent information derived from remotely-sensed images is used to infer relationships between the animal’s landscape preferences and the inferred vegetation characteristics.

These layers of inference introduce uncertainty, which may be reduced by directly correlating herbivore preferences based on GPS monitoring of herbivore movement with their landscape preferences. This approach reduces the uncertainty associated with the inference methods and removes the need to obtain ground-based vegetation calibration data. Wireless sensor networks enable high temporal-frequency GPS monitoring of animal locations to be directly linked to the spatially extensive measurements from remotely-sensed satellite images. An additional advantage of using WSNs is that no direct user involvement is required to download data from the devices, as is the case with traditional data loggers mounted on animals, and the data are streamed to the user in real-time. Studies that have combined multiple sensors within an integrated environment are rare and reflect the technological constraints of integration.

Wark and others [8] showed preliminary work on how ground-based multi-spectral sensors and satellite remotely-sensed data may be combined using a WSN. Bro-J?rgensen and others [32] showed how satellite-derived NDVI could be used to explain ranging patterns in antelope behaviour.Radio-transceivers and passive radio frequency identification (RFID) devices have been used to record information on animal ID and more recently to explore social interactions [33]. In particular, transceivers worn by a pair of animals can collect Cilengitide information on social encounters. The devices, referred to as contact or proximity loggers, record the date, time and duration of a close encounter. The inter-animal distance that is recorded as an encounter by the proximity logger can be adjusted by varying the transmission power setting of the device. Proximity loggers have been used to explore social interactions between cows and calves and also to explore potential risks of disease transmission by recording contacts between wild and domesticated animals [14,34].