Amphibian and reptile road-kills on tertiary roads in relation to landscape structure: using a citizen science approach with open-access land cover data

Table 6 Conditional probabilities of road-killed grass snakes for all land cover classes [P(E|B)]

CLC code	Land cover class	H(E)	H(E1)	H(E)+ H(E1)	P(E∩B)	P(B)	P(E\|B)
411	Inland marshes	0.143	0.126	0.269	0.001	0.001	0.531
142	Sport and leisure facilities	0.147	0.172	0.319	0.001	0.002	0.459
242	Complex cultivation patterns	2.191	4.501	6.692	0.011	0.034	0.330
311	Broad-leaved forest	1.296	3.849	5.145	0.007	0.026	0.251
211	Non-irrigated arable land	11.749	46.471	58.220	0.060	0.299	0.202
221	Vineyards	12.197	56.143	68.340	0.063	0.351	0.178
313	Mixed forest	0.380	1.851	2.232	0.002	0.011	0.174
243	Land principally occupied by agriculture, with significant areas of natural vegetation	0.900	4.362	5.262	0.005	0.027	0.172
231	Pastures	0.393	2.497	2.891	0.002	0.015	0.136
312	Coniferous forest	0.286	2.408	2.694	0.001	0.014	0.106
112	Discontinuous urban fabric	4.318	36.792	41.110	0.022	0.210	0.105
111	Continuous urban fabric	0.000	1.188	1.188	0.000	0.006	0.000
121	Industrial or commercial units	0.000	0.552	0.552	0.000	0.003	0.000
324	Transitional woodland shrub	0.000	0.088	0.088	0.000	0.000	0.000

Proportion of each land cover class per section containing road-killed grass snakes [H(E)] or not [H(E1)]. The probability of a road-killed grass snake on a certain land cover class [P(E∩B)] divided by the overall availability of this land cover class in the study area [P(B)] results in the conditional probability P(E|B). Land cover classes in descending order of P(E|B); the higher the P(E|B), the higher the probability of a road-kill on the specific land cover class

ISSN: 1472-6785