Comparing Factorial Least-cost Path Result from UNICOR and CoLa in Mapping Ecological Connectivity Networks in Mesoamerican Cloud Forests

Human-caused habitat loss paired with degradation driven by land-use and climate change represent the most challenging threats to global biodiversity. These factors increased habitat fragmentation and disrupted ecological processes. Characterizing landscape connectivity is necessary for conservation planning and maintaining viable populations under increasing anthropogenic pressure. Factorial least-cost path (FLCP) analysis is a functional connectivity approach that can infer the distribution and abundance of movement pathways between all core areas or individuals across a landscape. Unlike pairwise least-cost paths, FLCPs compute cost-distance paths for every source–destination pair, creating a spatial surface that reflects movement abundance and redundancy on pathways. FLCP analyses can be valuable for identifying redundant pathways and bottlenecks under changing environmental conditions yet require advanced technical expertise and computational capacity, limiting accessibility for non-specialist users.

This study compares two connectivity modeling frameworks: the Universal Corridor Network Simulator (UNICOR) and the Connecting Landscapes (CoLa) decision-support system to evaluate their performance when applied to Tropical Montane Cloud Forests (TMCF) landscapes. TMCFs are defined by persistent cloud cover that creates distinct climatic conditions necessary to maintain biodiversity. TMCFs are one of the most climate-sensitive habitats because of their restricted climate, natural fragmentation, and endemic levels. 

Both tools implement resistance-based approaches to modeling movement though there are differences in conceptual design, computational structure, and connectivity outputs. UNICOR applies a factorial least-cost path approach in which individual-based movement probabilities are computed for each source–destination pair, yielding a connectivity network appropriate for analyzing the overall permeability of the study regions landscapes. In contrast, CoLa functions as an integrated conservation decision-support system for conserving biological diversity, applying habitat suitability and resistance connectivity models simultaneously. Resulting in CoLa producing multiple connectivity maps including dispersal kernels, corridor areas, and least-resistance pathways, maintaining its focus on decision support in the field of biological conservation.

The growing demand for connectivity modeling outputs that are both interpretable and transparent fueled the inspiration for this comparison. Although UNICOR works better to represent emergent connectivity patterns across entire populations, CoLa emphasizes accessibility for assessing the trade-offs of development and biodiversity conservation. Despite their growing use, few studies have systematically compared outputs from these tools within the same ecological system, especially the newborn system CoLa. Users and practitioners have limited guidance for selecting connectivity approaches suited to spatial scales and data constraints.

To make these comparisons ecologically meaningful,  Resplendent Quetzal (Pharomachrus mocinno), Violet Sabrewing (Campylopterus hemileucurus), and Baird’s Tapir (Tapirus bairdii) were employed to initiate resistance surface development. These species span canopy, mid-story, and terrestrial movement strategies, enabling assessment of how each modeling framework represents connectivity across ecological guilds.

Both UNICOR and CoLa were implemented on the same resistance data derived from climate, vegetation productivity, topography, and anthropogenic pressure across montane regions throughout Mesoamerica. Resulting connectivity networks were compared in terms of spatial patterning, corridor density, bottleneck identification, and interpretability for conservation planning. Through this comparison, the study highlights how tool choice influences connectivity interpretation and conservation conclusions in montane environments.