MulTyLink (Multiple Types Linkages)

MulTyLink is an open source application, licensed under the GPL v3, designed to select connectivity linkages for distinct types of habitats, under a cost-efficient protocol. Since areas that can be used as linkages for one type of habitats may be barriers for other types, MulTyLink implements methods to optimise the selection of linkages free of barriers for every type of habitat. MulTyLink was conceived as a decision-support tool to be used in spatial conservation planning. The software provides users the flexibility to assign costs, friction values and habitat-specific barriers as input data. Based on these scores MulTyLink retrieves a network of linkages which may be visualised as a whole solution or independently for each habitat type.  

Why MulTyLink is needed for?

Fragmentation of natural habitats ranks among the more pervasive threats to biodiversity persistence [1-2]. Different species respond differently to fragmentation, depending on their ecological traits, phylogenetic history or geographical distributions. Therefore barriers affecting movements of some species may be valuable to assist movements of some other species. Moreover, distinct species may have different dispersal capabilities. In face of this, species-specific linkage areas should be selected to connect the habitats where the particular species occur.   

Typically, the selection of such conservation areas conflicts with competing land-uses and therefore an efficient set of areas should be part of final conservation proposals. To achieve efficiency, MulTyLink incorporates two optimisation procedures. One giving better (more optimal) solutions, with longer running times than the other. This gives to MulTyLink the flexibility to solve small-, medium- and large size problems.

Up to date a small set of freeware applications dedicated to the optimised selection of linkage areas exist. However these approaches consider the connectivity problem as one where all the natural habitats under analysis need to be linked (if possible) altogether. MulTyLink brings a novelty to the field by generalising the linkage optimization for problems where distinct sets of habitats exist, and linkages are needed to connect habitats within each set.

[1] Brooks, T.M., Mittermeier, R.A. et al. 2002. Habitat loss and extinction in the hotspots of biodiversity. Conservation Biology, 16: 909–923.

[2] Hanski, I. 2005. The shrinking world: Ecological consequences of habitat loss. Excellence in Ecology, 14. International Ecology Institute, Oldendorf/Luhe, Germany.

Major features of MulTyLink

• Easy-to-use, graphical interface with button and menu-driven commands. 
• Options to display friction and habitat-specific barriers for distinct combinations of habitat types.
• Tools for importing and visualizing empirical richness patterns, based on midpoint-range data.
• Capacity to implement two heuristics that adjust differently to problem structures.
• Export results to text files for analysis within GIS software, spreadsheets or statistics applications.
• MulTyLink is free. You can use it under the terms of the GPL v3 license. If you find any bugs or make any enhancements to the program please consdider contributing them back to the authors.

How to Cite MulTyLink

If you appreciate the effort that has gone into MulTyLink, please credit the application and the authors in any published work that makes use of results from MulTyLink, giving the MulTyLink website address and using the following preferred reference:

Brás, R., Cerdeira, J.O., Alagador, D. and Araújo, M.B., Linking habitats for multiple species, Environmental Modelling & Software, 40:336-339, 2013


We provide a user’s manual that includes detailed descriptions of MulTyLink’s main achievements, input data requirements, map visualisation capabilities and output files. The manual uses a case study to illustrate the use of MulTyLink.

User manual


1) Climatic classes of protected areas in Iberian Peninsula
Input data consists of a 100x100 1km-cell window respecting a fraction of Iberian Peninsula, where two climatic types of protected areas (PAs) are located. Barriers for each PA type are the unprotected cells with the highest dissimilar climates from the climate characterizing that PA type. A cost value of 1 was given to each of the unprotected cells along with a friction value respecting the human footprint index (from 0 to 100, with higher values characterizing high human disturbance). PA cells were given cost and friction 0.

Based on this cost setting, the problem is one of minimising the total number of selected cells to link (if barriers and friction allow) all the PA of the same type.

For more details regarding data see Alagador, D., Triviño, M., Cerdeira, J.O., Brás, R., Cabeza, M. and Araújo, M.B. Linking like with like: optimising connectivity between environmentally-similar habitats, Landscape Ecology, 27, 291-301, 2012.

fig 2

2) Three reptile species in Iberian Peninsula protected areas
Input data consists of 2310 10' squared cells (aprox. 16 km edge length) located in the continental Iberian Peninsula and Balearic Islands. The habitats to be linked are the cells where three threatened reptile species (Lacerta bilineata, Lacerta schreiberi and Coronella austriaca) occur and where more than 75% of the cell's surface contains protected areas. Barriers for each species represent the 50% least suitable cells (in respect to climate), obtained after using an ensemble of species distribution models with climatic predictors. A maximum permeability (0 friction value) was assigned to each cell along with a cost value representing the percentage of cell's surface not listed in Natura 2000.

In this way the problem turns into one of maximising the selection of linkage areas in cells with high Natura 2000 coverage. 

For more details regarding data see Brás, R., Cerdeira, J.O., Alagador, D. and Araújo, M.B. Linking habitats for multiple species, Environmental Modelling & Software, in press, DOI: 10.1016/j.envsoft.2012.08.001

fig 3

Download 2.0.2

Windows 32-bits executable
Linux (Fedora 15) 64-bits executable
Source code
Examples data files

To compile the program for other architectures please read the file README.TXT in the source package.