Habitat selection by the Spiny-tailed lizard (Uromastyx aegyptia): A view from spatial analysis

Abstract

Many factors affect the habitat selection for animal species, which in turn may greatly affect their distribution in different ecosystems. Understanding the processes that affect habitat selection is also critical for guiding and managing conservation initiatives. Our study aimed to assess the habitat selection by free-ranging Spiny-tailed lizard (Uromastyx aegyptia) by analyzing a geospatial data connecting its burrow parameters to different habitat characteristics within selected sites in Hail region, Saudi Arabia. We examined evidence and patterns of significant spatial clustering for (366) active burrows by linking their parameters (burrow entrance size, burrow entrance width and burrow entrance height), their reference geographical locations and, two habitat characteristics defined by soil type and vegetation cover. The objective of the analysis was to increase the understanding on the burrows aggregation process in the space and, to describe its possible relation to other spatial habitat configurations. Analysis of distances based on the Nearest Neighbor Index (NNI) and hotspots detection in Nearest neighbor hierarchical clustering (Nnh) suggested twelve (12) spatial clusters located within the study area. In addition, a spatial ordinary least square (OLS) and Poisson regression models revealed significant effects of soil type and vegetation cover on burrow parameters (OLS, p < 0.05; Poisson, p < 0.001), which indicate a strong association between burrows parameters and habitats characteristics. Findings from the study also suggest that other factors such as elevations, highways, and human settlement concentration spots could possibly play a major role in defining burrow spatial aggregation and furthermore have a significant impact on habitat selection.

Keywords: Burrow parameters; Distance analysis; Hail region; Spatial regression; Uromastyx aegyptia.

Fig. 1

Study areas and burrows locations in Hail region, Saudi Arabia.

Fig. 2

(a) Moran scatter plot and (b) reference distribution for Moran’s I.

Fig. 3

LISA (a) cluster and (b) significant maps (Significance level is shown as *P < 0.05, **P < 0.01).

Fig. 4

Clusters as revealed by (a) Nearest Neighbor Hierarchical Clustering (Nnh) method and, the burrow clustering based on (b) soil type and (c) vegetation cover as determined by nearest neighbor analysis.

Fig. 5

A map showing the burrow locations within the study area and across (a) different elevation values, (b) roads and highways and (c) towns and villages.