From theory to reality: validating connectivity models with on-the-ground movement data

Donaghy's Corridor

Protected areas, such as Donaghys Corridor Nature Reserve in northeastern Australia, play an important role in providing networks of connectivity.

Enhancing landscape connectivity is one of the most important tools we have for enabling species to traverse increasingly human-dominated landscapes. Given this importance, conservation researchers and practitioners have developed a wide range of approaches for mapping and measuring connectivity.

Despite this proliferation of approaches, what’s often missing is solid, empirical evidence that species use pathways that we humans flag as important for connectivity. Validating connectivity models with on-the-ground movement data and clarifying the role that protected areas (PAs) play in connectivity will be particularly important as species shift to track suitable climatic conditions.

A new study seeks to do just that by asking: When the rubber hits the road (or rather, when the paws hit the trail) how well does species actual movement correspond to connectivity models?

Using high-resolution data from 10 GPS-collared fisher (Pekania pennanti) in a central Alberta biosphere reserve, researchers compared speed (step-length) and linearity of movement (turn angle) to three traditional connectivity mapping schemes. They created test models using proximity to and density of different land cover types (e.g., coniferous forest, grasslands, cropland, and waterbodies) to represent 1) a corridor scheme, 2) a least-cost path scheme, and 3) a stepping-stone scheme that leverages protected areas across a working landscape.

Pacific Fisher

The Fisher (Pekania pennanti) is a member of the weasel family and common across Canada and some parts of the northern United States. (USFWS)

The researchers hypothesized that idealized movement under each scheme would look different. For example, if fishers were using stepping stones, they would likely dwell and move freely within PAs (corresponding to low speed and highly complex paths) while traversing highly modified areas quickly and directly.

Based on the telemetry data, the fishers’ actual movement was best explained by the corridor scheme. In other words, the animals tended to move among structurally similar natural features across the landscape and did not appear to preferentially travel to or through PAs. In general, we know that PAs are critical for species persistence, but results like these underscore the importance of considering the matrix between delineated PAs in conservation priorities.

This study area happens to have a fair amount of natural vegetation on private lands beyond PAs, which contributed to landscape permeability in general, and may mean that the corridor-like movement approximates least-cost movement quite well, too. Plus, using speed and linearity of movement as evidence for a particular connectivity scheme may not hold in all landscapes or for all species. Nevertheless, this study highlights that we would be wise to use “animal-defined” corridors in prioritizing areas for enhancing or protecting connectivity whenever possible—and that our theoretical models ought to be rigorously validated with on-the-ground data.


Stewart, F. E., Darlington, S., Volpe, J. P., McAdie, M., & Fisher, J. T. 2019. Corridors best facilitate functional connectivity across a protected area network. Scientific Reports 9(1): 10852.

Calabrese, J. M., and Fagan, W. F. 2004. A comparison‐shopper’s guide to connectivity metrics. Frontiers in Ecology and the Environment 2(10): 529-536.

Di Marco, M., Ferrier, S., Harwood, T. D., Hoskins, A. J., and Watson, J. E. 2019. Wilderness areas halve the extinction risk of terrestrial biodiversity. Nature 573: 582-585.

Littlefield, C. E., Krosby, M., Michalak, J. L., and Lawler, J. J. 2019. Connectivity for species on the move: supporting climate‐driven range shifts. Frontiers in Ecology and the Environment 17(5); 270-278.

Rayfield, B., Fortin, M. J., and Fall, A. 2011. Connectivity for conservation: a framework to classify network measures. Ecology 92(4): 847-858.

2019-10-09T08:48:26-04:00 October 9th, 2019|

About the Author:

Caitlin Littlefield
Caitlin Littlefield is a Northwest Climate Adaptation Science Center research fellow at the University of Montana. Her research focuses on strategies to maintain landscape connectivity and promote forest recovery from disturbance under climate change. She received her PhD in landscape ecology from the University of Washington, her MS in forest ecology from the University of Vermont, and a BA from Middlebury College.