Density, occupancy and detectability of tortoise species (Chelonoidis spp.) in the Atlantic Forest: implications for conservation and management
Author
Pereira-Ribeiro, Juliane
Author
Ferreguetti, Atilla Colombo
Author
Bergallo, Helena
Author
Rocha, Carlos Frederico Duarte
text
Journal of Natural History
2024
2024-09-02
58
37 - 40
1529
1545
http://dx.doi.org/10.1080/00222933.2024.2390207
journal article
10.1080/00222933.2024.2390207
1464-5262
13758587
Chelonoidis carbonarius
We recorded 20 individuals of
C. carbonarius
, all detected during transects. We detected more individuals in the northern region of the VNR (70%). Most individuals of
C. carbonarius
were recorded in the coastal plain forest (95%), at an average distance of
329.9 m
from the nearest water resource (on average, 17.8% farther away from water courses than
C. denticulatus
).
Figure 2.
Relationship between the probability of detection of
Chelonoidis denticulatus
(A) and
Chelonoidis carbonarius
(B) with the accumulated rainfall of the day before sampling (previous rainfall) in the Vale Natural Reserve, Brazil.
Table 1.
Single-species, single-season occupancy and detectability models for the
Chelonoidis denticulatus
and
C.
carbonarius
in the Vale Natural Reserve, Brazil. Covariates: percentage of forest cover in a 200 m buffer (%; ‘forest cover’), distance from water resource (in metres; ‘water’), distance from natural grasslands (in metres; ‘grassland’), accumulated rainfall from the day before sampling (in millimetres; ‘prev_rain’). Ψ = occupancy, p = detectability, AIC = Akiake information criterion, AICcw = Akaike weight, K = number of parameters.
| Model |
AICc |
ΔAICc |
AICcw |
K |
|
Chelonoidis denticulatus
|
| Ψ (forest, water), p (prev_rain) |
203.27 |
0.00 |
0.41 |
5 |
| Ψ (water), p (prev_rain) |
203.78 |
0.51 |
0.32 |
4 |
| Ψ (.), p (prev_rain) |
205.80 |
2.53 |
0.12 |
3 |
| Ψ (forest), p (prev_rain) |
206.98 |
3.71 |
0.06 |
4 |
| Ψ (grassland, water), p (prev_rain) |
208.16 |
4.90 |
0.04 |
5 |
| Ψ (forest, grassland, water), p (prev_rain) |
208.43 |
5.16 |
0.03 |
6 |
| Ψ (grassland), p (prev_rain) |
210.03 |
6.76 |
0.01 |
4 |
| Ψ (forest, grassland), p (prev_rain) |
211.61 |
8.35 |
0.01 |
5 |
| Ψ (.), p (.) |
214.09 |
10.83 |
0.00 |
2 |
|
Chelonoidis carbonarius
|
| Ψ (water), p (prev_rain) |
148.67 |
0.00 |
0.53 |
4 |
| Ψ (.), p (prev_rain) |
151.17 |
2.50 |
0.15 |
3 |
| Ψ (forest, water), p (prev_rain) |
151.18 |
2.51 |
0.15 |
5 |
| Ψ (grassland, water), p (prev_rain) |
153.04 |
4.37 |
0.06 |
5 |
| Ψ (forest), p (prev_rain) |
154.05 |
5.38 |
0.04 |
4 |
| Ψ (forest, grassland, water), p (prev_rain) |
154.53 |
5.86 |
0.03 |
6 |
| Ψ (grassland), p (prev_rain) |
154.75 |
6.08 |
0.03 |
4 |
| Ψ (forest, grassland), p (prev_rain) |
157.70 |
9.03 |
0.01 |
5 |
| Ψ (.), p (.) |
157.91 |
9.24 |
0.01 |
2 |
We detected the species at 11 of 24 sampling sites. The estimated population size of
C. carbonarius
for the entire area was 105 (105 ± 21.10) individuals. The estimated probability of occupancy was 59% (0.59 ± 0.15) and the estimated detection probability was 8% (0.08 ± 0.02).
The detectability of
C. carbonarius
was positively influenced by the accumulated rainfall of the previous day of sampling (
Figure 2
). The occupancy of
C. carbonarius
was influenced only by the distance from the water, presenting the highest probabilities of occupancy in areas closer to the water (
Figure 3
).