To deal with increasing demand for multi-taxa, multi-scaled connectivity maps, researchers have created methods such as discretizing the landscape (i.e., ‘tiling’) or using multiple moving windows in order solve large connectivity problems. Meanwhile, the use of Circuitscape, one of the most commonly utilized connectivity software programs over the last few years, has expanded from predicting gene flow across landscapes to predicting other ecological phenomenon such as the spread of wildfire and invasive species. However, existing software for ecologists has always been limited computationally, and researchers have often been forced to coarsen their data, ask smaller questions, or create techniques similar to those mentioned above to overcome these limitations.
In light of these theoretical extensions that employ circuit-theory based connectivity solutions across time and space, we created the software GFlow. The software is designed for use on supercomputers or high performing desktop computers running modern versions of Linux or Apple’s OS X, and massively parallelizes circuit-theory calculations.
Using this software, we have been able to solve problems formerly estimated to require > 2.5 years of compute time in under 1 week. We suggest this software will allow researchers to ask bigger questions from their data and continue exploring new ways to use circuit-theory connectivity. Perhaps more importantly, GFlow can facilitate a deeper exploration into the effects of model inputs, scaling trade-offs, and multi-scaled scenarios across large areas.
GFlow website: github.com/Pbleonard/GFlow
Leonard, P.B., E. B. Duffy, R. F. Baldwin, B. H. McRae, V. B. Shah, and T. K. Mohapatra. 2017. GFlow: software for modelling circuit theory-based connectivity at any scale. Methods in Ecology and Evolution. DOI: 10.1111/2041-210X.12689