Distribution, habitat and vulnerability to climate change of the Endangered Leptopelis xenodactylus

Figures & data

Figure 1. Distribution records for Leptopelis xenodactylus collected prior to (magenta squares) and during (black dots) this study. The areas of average relative suitability for the species (coloured polygons) were obtained from the output of the first distribution model. Inset: the mask area (red; see text) for Leptopelis xenodactylus within the KwaZulu-Natal province (grey area) of South Africa.

Table 1. Variables included in the distribution models according to their frequency of occurrence at locations (n = 48) with Leptopelis xenodactylus.

Variable	Frequency
Vegetation type
Temperate Alluvial Wetlands	22
Mooi River Highland Grassland	10
Drakensberg Wetlands	8
Drakensberg Foothills Moist Grassland	3
Midlands Mistbelt Grassland	2
uKhahlamba Basalt Grassland	2
Eastern Temperate Wetlands	1
Landform type
U-shaped valley	26
Upper slopes, mesas	8
Canyons, deeply incised streams	7
Upland drainages, headwaters	1
Open slopes	1
Mountain tops, high ridges	3
Plains	2
Elevation (m a.s.l.)
1 000–1 300	4
1 301–1 600	17
1 601–1 900	25
1 901–2 000	2
Mean summer monthly potential evaporation (mm)
130–140	5
141–150	12
151–160	31
161–170	1
Variable	Frequency
Average summer mean daily maximum temperature (°C)
21	4
22	8
23	28
24	3
25	4
26	1
Average summer median monthly rainfall (mm)
90–110	8
111–130	30
131–150	9
151–170	1
Average summer mean daily average relative humidity (%)
71–73	4
74–76	19
77–79	20
80–82	5
Plant available water (mm)
30–50	1
51–70	8
71–90	21
91–110	12
111–130	5
131–150	1

Figure 2. Jackknife test of variable importance using training gain (top) and using test gain (middle), and jackknife test of variable importance using area under the receiver operator characteristic curve on test data. See the text for explanation of the acronyms.

Figure 3. Average relative suitability of areas for Leptopelis xenodactylus within the predicted distribution range.

Table 2. Variables included in the MaxEnt environmental niche model for Leptopelis xenodactylus. VEG = vegetation type, RH = average summer relative humidity (%), RAIN = average summer median monthly rainfall (mm), PEVAP = mean summer monthly potential evaporation (mm), LF = landform, PAW = soil plant available water (mm), DEM = elevation (m a.s.l.), MXTEMP = average summer mean daily maximum temperatures (°C). Summer was defined as September to March. See Material and Methods for further explanation of the variables.

Variable	Percent contribution	Permutation importance
VEG	57.3	41.9
RH	14.8	0.3
RAIN	12.7	13.1
PEVAP	6.2	33.0
LF	4.9	6.8
PAW	2.2	0.2
DEM	1.3	4.5
MXTEMP	0.6	0.2

Table 3. Localities (n = 41) of Leptopelis xenodactylus in relation to the vulnerability framework and the two downscaled climate models (HadCM2 and GFDL2.1; Jewitt et al. 2015a).

	Climate models: HadCM2 (H), GFDL2.1 (G)
	Robust	Susceptible	Constrained	Vulnerable
No. of localities	H: 8 G: 32	H: 24 G: 0	—	H: 9 G: 9
Percentage	H: 19.5 G: 78.0	H: 58.5 G: 0	—	H: 22.0 G: 22.0

Jewitt D, Erasmus BFN, Goodman PS, O’Connor TG, Hargrove WW, Maddalena DM. 2015a. Climate-induced change of environmentally defined floristic domains: A conservation based vulnerability framework. Appl Geogr. 63:33–42. https://doi.org/10.1016/j.apgeog.2015.06.004

 Web of Science ®Google Scholar