Each autumn about 30 wardens of the Gran Paradiso National Park (GPNP) conduct a ground count of the Park's ibex population by walking over established routes that include stops at fixed locations. Wardens visually search the ibex habitat in their assigned surveillance zones, which have an average area of about 10 km2. Each warden is familiar with the census area and with the habits of ibex within that area. During the census, most ibex are in open habitats above the timberline, which greatly aids in their detection and enhances our confidence in the reliability of the observations. The census of the entire park is conducted over two consecutive days in September. Ibex are observed with binoculars and spotting scopes and classified as kids, yearlings, adult males or adult females. Park wardens report that on census dates, the weather was generally clear with little or no snow cover.
The reliability of the GPNP census is currently being studied, and only limited information about its validity is available. In 2000, a sample of marked ibex was available within one of the census units. That July, wardens saw 28 of 32 marked ibex (87.5%). We suggest that the combination of experienced observers, fixed census routes and open terrain has provided population estimates with little or no bias.
An important limitation of our data is the difficulty of estimating confidence limits on yearly ibex counts. The wardens of the GPNP normally perform two ibex censuses a year, one in the late spring and one in the autumn. Beginning in 1986 the date of the spring census was changed from June (before parturition) to July (after parturition). This discontinuity of methodology makes spring census observations difficult to include in the simple models explored in this paper, but some conclusions about the reliability of the censuses are nonetheless possible. As reported in von Hardenberg et al. (2000), the spring and fall counts are highly correlated (total ibex: r = 0.95, adult males: r = 0.97, adult females: r = 0.95, yearlings: r = 0.91, kids: r = 0.92), with about 14% more ibex being counted in the fall due to their annual dispersion outside the park boundaries in spring. These results lead us to believe that the yearly counts are indeed fairly reliable.
A chief danger in this modeling exercise is that of a systematic bias in sightability, such as if ibex were more visible in the autumn following winters with relatively low snow accumulation. We have no reason to suspect such a bias, as the counts take place several months after the end of snow melt on days of generally fair weather, and the ibex are in open terrain above the tree line. We believe that the relationship between snow depth and population changes is robust since the data come from independent sources.
Table A1. Observed total ibex for the fall census in the Gran Paradiso National Park (Italy) and climate data at the nearby Serrù station. Daily meteorological observations were aggregated into seasonal quantities, with summer defined as May–October, winter as November–April, and spring as May–July. For variables that involve snow depth, winters were defined as November–May to account for a lingering snowpack.
Year N X Y SD SL1 SL2 TmaxW TminW TmaxS TminS PrecipSpr PrecipWin PrecipSum 1 1956 2767 7.93 0.0376 2 1957 2873 7.96 0.159 3 1958 3367 8.12 0.0504 4 1959 3541 8.17 -0.0495 5 1960 3370 8.12 0.0864 9.4 1.8 279.6 460.0 6 1961 3674 8.21 0.0395 158.2 163 27 0.3 -7.0 11.0 3.3 150.8 427.2 314.6 7 1962 3822 8.25 -0.31 216.4 200 78 -2.1 -9.7 10.2 2.7 232.6 440.4 482.2 8 1963 2803 7.94 0.139 142.7 108 22 0.9 -6.9 10.7 3.1 186.2 464.4 383.6 9 1964 3222 8.08 0.0933 97.2 66 0 -1.4 -8.5 9.3 2.1 248.6 175.2 648.0 10 1965 3537 8.17 -0.0797 154.2 129 51 -0.1 -7.9 9.8 2.4 187.9 447.3 656.3 11 1966 3266 8.09 0.0729 115.4 45 0 -1.2 -8.3 10.5 2.8 248.2 414.0 391.2 12 1967 3513 8.16 0.0843 77.8 47 0 0.3 -7.6 9.8 2.0 472.4 328.4 902.6 13 1968 3822 8.25 -0.177 181.5 148 43 -1.0 -9.1 9.9 2.1 424.0 697.2 683.4 14 1969 3203 8.07 0.0688 134.0 145 0 -2.9 -10.2 9.6 1.9 361.0 471.6 682.2 15 1970 3431 8.14 -0.0144 138.9 114 45 -1.1 -8.6 10.4 2.4 376.6 506.2 569.4 16 1971 3382 8.13 -0.208 273.5 159 119 -1.1 -8.0 8.1 0.3 335.2 1411.4 657.4 17 1972 2746 7.92 0.14 84.7 36 0 -1.5 -8.4 9.2 2.5 470.0 363.1 846.6 18 1973 3158 8.06 -0.0134 195.5 154 109 0.4 -6.0 9.7 1.8 387.2 733.9 612.0 19 1974 3116 8.04 0.142 139.3 97 64 0.3 -6.2 10.4 3.5 446.8 313.5 721.5 20 1975 3590 8.19 0.0447 130.0 90 26 1.4 -6.3 11.0 2.8 243.0 373.2 818.6 21 1976 3754 8.23 -0.443 279.6 212 192 -1.1 -8.1 9.5 1.7 720.6 1341.1 1247.0 22 1977 2410 7.79 0.247 198.3 139 127 0.3 -7.8 10.5 1.8 276.4 557.1 413.9 23 1978 3084 8.03 -0.0137 116.3 118 0 2.1 -7.8 11.4 1.5 180.5 515.2 741.9 24 1979 3042 8.02 0.0612 186.0 154 88 1.2 -9.3 9.5 -0.0 288.6 428.9 608.6 25 1980 3234 8.08 0.0536 71.3 48 10 -1.4 -9.5 7.8 -0.5 473.3 540.5 959.9 26 1981 3412 8.14 -0.0471 103.5 88 5 -0.8 -9.8 9.0 -0.1 216.6 369.9 684.0 27 1982 3255 8.09 -0.0211 153.1 136 20 -1.7 -10.0 8.5 -0.6 369.0 687.6 616.0 28 1983 3187 8.07 0.0535 103.5 83 14 -2.2 -11.6 7.9 -1.8 522.8 382.7 842.0 29 1984 3362 8.12 0.107 127.1 87 9 -1.8 -11.6 9.2 -0.7 499.0 546.1 636.3 30 1985 3740 8.23 0.0455 142.0 118 49 -1.9 -9.6 11.3 4.4 295.0 1302.3 587.6 31 1986 3914 8.27 0.0901 84.6 61 4 0.7 -7.6 10.0 2.2 325.2 388.4 798.5 32 1987 4283 8.36 0.00466 106.4 89 0 -0.6 -7.5 13.1 3.6 377.0 456.6 756.2 33 1988 4303 8.37 0.087 58.5 31 8 3.6 -6.3 13.8 2.8 99.2 460.6 210.1 34 1989 4694 8.45 -0.0135 45.9 10 1 2.2 -7.2 11.0 2.8 177.2 324.6 573.0 35 1990 4631 8.44 0.0338 139.9 104 42 -1.6 -8.9 9.4 2.3 169.4 698.6 598.0 36 1991 4790 8.47 -0.00754 36.3 0 0 -0.5 -7.1 9.0 2.7 492.2 221.5 1147.6 37 1992 4754 8.47 0.0486 93.7 45 13 -0.5 -6.9 8.3 2.0 299.0 589.6 764.8 38 1993 4991 8.52 -0.188 99.9 64 8 -1.8 -8.0 9.9 3.6 451.4 366.0 1011.2 39 1994 4136 8.33 0.0527 71.5 15 7 -0.2 -6.5 9.8 2.6 230.4 694.0 498.6 40 1995 4360 8.38 -0.0867 120.9 95 12 -0.9 -8.5 9.0 2.2 401.2 585.2 732.4 41 1996 3998 8.29 -0.11 146.3 130 47 0.3 -7.5 10.4 3.6 413.6 550.6 597.0 42 1997 3581 8.18 0.033 72.5 21 0 0.4 -6.7 10.1 3.1 297.0 437.0 536.2 43 1998 3701 8.22 -0.00732 64.1 15 0 -0.8 -8.0 9.7 3.6 326.2 245.4 817.2 44 1999 3674 8.21 -0.0115 42.9 2 0 -0.6 -5.7 11.0 4.3 432.8 605.0 45 2000 3632 8.2
Table A2. Variable names used in Table A1.
Abbreviation Year N X Y SD SL1 SL2 TmaxW TminW TmaxS TminS PrecipSpr PrecipWin PrecipSum
von Hardenberg, A., B. Bassano, V. Percacino, A. R. Jacobsen, J. von Hardenberg, and A. Provenzale. 2000. Preliminary analysis of the temporal variability of the Alpine ibex population in the Gran Paradiso National Park. Ibex Journal of Mountain Ecology 5:201–210.