Table 1 Models considered to investigate mortality by poaching and other causes, resighting and recovery probabilities of female (F) and male (M) deer within the Stelvio National Park, between 2008 and 2017. The table reports: model structure for mortality, resight and recovery probability; np number of parameters, Dev relative deviance, AICc Akaike Information Criterion corrected for sample size, ΔAICc the AICc difference between the current model and the model with the lowest AICc value; w = Akaike’s weight calculated for the candidate models (i.e. models fitted to investigate biological hypotheses on survival); Hypothesis: different hypotheses on the impact of poaching relative to other causes of mortality. Model notation: in ‘A’, numbers represent the age intervals; ‘t’ = temporal effects; ‘×’ and ‘+’ indicate interactive and additive effect, respectively; ‘,’ = different parameters were considered; ‘.’ = constant; ‘Poach’ = poaching; ‘Tag’ = ear tags; ‘Collar’ = optical collars and inactive GPS collars. All models assumed: a 2-year variation in the probabilities of losing GPS signal after deployment; constant probability of GPS drop-off; constant probabilities of determining causes of death. Hypothesis-driven models are in italics, models with the lowest AICc for either sex are in bold
From: Unravelling the sex- and age-specific impact of poaching mortality with multievent modeling
Model | Mortality | Resight | Recovery | np | Dev | AICc | ΔAICc | w | Hypothesis |
|---|---|---|---|---|---|---|---|---|---|
1 F | [Poach, Other] × A(1,2:7,8+) | [Tag, Collar] × t | t | 40 | 820.13 | 910.51 | 28.79 | ||
2 F | [Poach, Other] × A(1:3,4:7,8+) | [Tag, Collar] × t | t | 40 | 821.42 | 911.80 | 30.08 | ||
3 F | [Poach, Other] × A(1,2:3,4:7,8+) | [Tag, Collar] × t | t | 42 | 819.68 | 915.19 | 33.47 | ||
4 F | [Poach, Other] × A(1,2:7,8+) | [Tag, Collar] × t | . | 31 | 826.27 | 894.38 | 12.66 | ||
5 F | [Poach, Other] × A(1,2:7,8+) | [Tag, Collar] + t | . | 24 | 835.38 | 887.00 | 5.28 | ||
6 F | [Poach, Other] × A(1,2:7,8+) | t | . | 23 | 835.38 | 884.70 | 2.98 | 0.106 | Mortalities vary interactively with age |
7 F | [Poach, Other] × A(1,2:7,8+) | [Tag, Collar] | . | 15 | 858.43 | 889.84 | 8.12 | ||
8 F | [Poach, Other] × A(1,2:7,8+) | . | . | 14 | 858.45 | 887.67 | 5.95 | ||
9 F | [Poach, Other] + A(1,2:7,8+) | t | . | 21 | 837.15 | 881.91 | 0.19 | 0.426 | Mortalities vary in parallel with age |
10 F | [Poach] vs Other × A(1,2:7,8+) | t | . | 21 | 836.96 | 881.72 | 0.00 | 0.469 | Poaching mortality is constant |
1 M | [Poach, Other] × A(1,2:7,8+) | [Tag, Collar] × t | t | 39 | 498.68 | 607.88 | 45.75 | ||
2 M | [Poach, Other] × A(1:3,4:7,8+) | [Tag, Collar] × t | t | 39 | 492.34 | 601.54 | 39.41 | ||
3 M | [Poach, Other] × A(1,2:3,4:7,8+) | [Tag, Collar] × t | t | 41 | 492.14 | 609.29 | 47.16 | ||
4 M | [Poach, Other] × A(1:3,4:7,8+) | [Tag, Collar] × t | . | 31 | 503.03 | 583.40 | 21.27 | ||
5 M | [Poach, Other] × A(1:3,4:7,8+) | [Tag, Collar] + t | . | 24 | 510.54 | 568.97 | 6.85 | ||
6 M | [Poach, Other] × A(1:3,4:7,8+) | t | . | 23 | 514.66 | 570.18 | 8.05 | ||
7 M | [Poach, Other] × A(1:3,4:7,8+) | [Tag, Collar] | . | 15 | 530.44 | 564.31 | 2.18 | 0.251 | Mortalities vary interactively with age |
8 M | [Poach, Other] × A(1:3,4:7,8+) | . | . | 14 | 536.32 | 567.68 | 5.55 | ||
9 M | [Poach, Other] + A(1:3,4:7,8+) | [Tag, Collar] | . | 13 | 533.24 | 562.13 | 0.00 | 0.744 | Mortalities vary in parallel with age |
10 M | [Poach] vs Other × A(1:3,4:7,8+) | [Tag, Collar] | . | 13 | 543.16 | 572.04 | 9.91 | 0.005 | Poaching mortality is constant |