Kumpula.indd 225Kumpula 2003: Polar Research 22(2), 225–233 It is a commonly accepted tenet that the popula- tion dynamics of ungulates in predator-free envi- ronments are strongly infl uenced by a combina- tion of stochastic variation and population density (Messier 1991; Sæther 1997; Portier et al. 1998; Milner et al. 1999; Aanes et al. 2000). As a con- sequence of stochastic variation, a fi xed, stable equilibrium between herbivores and their food plants is unlikely to exist in natural grazing sys- tems (Ellis et al. 1991; Behnke & Scoones 1992; Aanes et al. 2000). However, it has also been observed that population size can oscillate within certain limits in many herbivore populations. Therefore it can be supposed that within a longer period population density can fl uctuate around some mean density with a certain variance (Mur- dock 1994; Turchin 1995; Sæther 1997). It can be also supposed that in reindeer (Rang- ifer t. tarandus) and caribou (R. t. groenlandicus) populations the main source of stochastic varia- tion is of climate-mediated origin. Thus, knowl- edge of the effects of various weather and snow Effects of weather and snow conditions on reproduction and survival of semi-domesticated reindeer (R. t. tarandus) Jouko Kumpula & Alfred Colpaert In this work we investigated the effects of local weather and snow con- ditions on the reproduction and survival of semi-domesticated reindeer (Rangifer t. tarandus) from 1962 to 1987 in four separate study areas in northern Finland. Reindeer density had no negative effect on calf percent- age in slaughter (October–January) in any area. The higher the number of very warm days or mean temperatures in summer, the lower was the calf percentage in all forested areas (Sodankylä, Muonio and Ivalo). In the Muonio area, calf percentage was also reduced by winters with abundant snow accumulation. In the open, mountainous Kevo area, calf percentage was reduced with a high number of warm days (mean T > 0 °C) during the previous December. However, if the previous May was warm in that area, it had an opposite effect. High reindeer densities seemed to increase mor- tality only in the Sodankylä area. Abundant snow accumulation during winter reduced the survival index of reindeer both in the Sodankylä and Ivalo areas. In the Muonio area, mortality of reindeer was increased with a high number of warm days in December. In the Kevo area, reindeer density or any climatic parameters could not explain yearly differences in mortality. This study indicated that permanent changes in climatic conditions could have different impacts in woodland as opposed to open, mountainous regions. In general, if global climatic change means weather instability in early winter and more snow, it also brings more diffi culties for reindeer. However, if climate change means that snow melts earlier in the spring, some conditions could become more favourable for reindeer. J. Kumpula, Reindeer Research Station, Finnish Game and Fisheries Research Institute, FI-99910 Kaa- manen, Finland, jouko.kumpula@rktl.fi ; A. Colpaert, Dept. of Geography, University of Oulu, FI-90014 University of Oulu, Finland. 226 Effects of weather and snow conditions on reproduction and survival of reindeer conditions on the demography of reindeer and caribou populations helps us to understand better the fl uctuations of productivity and animal densi- ty in these populations. However, if or when sto- chastic variation is bounded into certain limits, animal density and pasture condition are impor- tant measures when trying to optimize pasture use and productivity of reindeer or caribou herds in a long run (Klein 1968; Skogland 1985, 1986; Ouellet et al. 1997; Kumpula et al. 1998; Fynn & O’Connor 2000; Kumpula, Colpaert et al. 2000; Kumpula 2001a; Moxnes et al. 2001). There is also an interaction between popula- tion density and climatic variables, i.e. the effect of climate is stronger at high than at low reindeer densities (Aanes et al. 2000). This is probably because at high animal densities pastures tend to deteriorate and the intensity of direct food com- petition increases (Klein 1968; Leader-Williams et al. 1987; Henry & Gunn 1991; Manseau et al. 1996; Augustine et al. 1998; Crête & Doucet 1998; Alpe et al. 1999; Kumpula, Colpaert et al. 2000). Especially diffi cult snow conditions in winter can increase mortality and reduce the reproduction rate in genus Rangifer (Klein 1968; Helle 1980; Reimers 1982; Adamczewski et al. 1988; Kumpu- la & Nieminen 1992; Adams & Dale 1998; Aanes et al. 2000). Another factor explaining much of the yearly variation in body weights of reindeer and caribou in autumn, and also affecting calf mortality, is the level of insect harassment, which has been found to correlate with daily tempera- tures and wind speed in summer (Helle & Tar- vainen 1984; Mörschel & Klein 1997; Mörschel 1999). A large number of parasitic insects affects reindeer and caribou mainly by decreasing graz- ing and increasing energy expenditures. The objective of this work was to study the effects of local weather and snow conditions on the reproduction and survival of semi-domesti- cated reindeer (R. t. tarandus) in four areas in northern Finland. Vegetation, pastures, topog- raphy and climatic conditions differ among the study areas. One important aspect was there- fore to look at which weather effects were the most detrimental in each area. Reproduction and sur vival of reindeer were studied before 1987 because since then supplementary feeding of rein deer has improved the condition of reindeer, masking the effects of adverse weather conditions (see Kojola et al. 1995; Kumpula et al. 2002). Material and methods Weather and snow data from 1961 to 1987 were obtained from four stations of the Finnish Mete- orological Institute (Sodankylä, Muonio, Ivalo and Kevo) in nothernmost Finland (Fig. 1). The following parameters for each month were used: mean temperature, maximum temperature, mini- mum temperature, total precipitation (mm), snow depth on the 15th day, number of days when the mean temperature exceeded 0 °C, 5 °C and 15 °C, and number of days when the maximum tem- perature was higher than 25 °C. Also calculated were mean temperature and total precipitation for January–February, March–April, June–August, September–October and November–December. For the whole summer period (June–August), the following parameters were calculated: the number of days when the mean temperature exceeded 15 °C and the number of days when the maximum temperature was higher than 25 °C. The sum of monthly snow depths was calculat- ed for November–December, January–February, Fig. 1. The Finnish reindeer management area and the four separate study areas. Location of the stations of the Mete- orological Institute in the study areas have been marked. Reindeer management districts in Kevo are (1) Paistunturi and (2) Kaldoaivi; in Ivalo (3) Hammastunturi and (4) Ivalo; in Muonio (5) Muonio and in Sodankylä (6) Sattasniemi and (7) Oraniemi. 227Kumpula 2003: Polar Research 22(2), 225–233 March–April and then November–April. The dif- ference in mean temperature between December and November was also calculated. Calf percentage (calves per 100 females) in round-ups (October–January) and the mortality index of reindeer and reindeer density (reindeer/ km² land area, winter stock) were calculated for each year from the statistics of the Finnish Rein- deer Herders’ Association in the four study areas (Fig. 2). The survival index of reindeer was calcu- lated comparing the number of adult (over 1 year old) reindeer counted in round-ups to the number of all reindeer left alive (adults + calves) in the previous year. The survival index was therefore the combination of natural mortality and count- ing accuracy of reindeer so that both of these ele- ments affected the size of counted reindeer popu- lation in a certain year compared to the previous year. Although counting accuracy of reindeer has varied between years, the survival index still refl ects a certain level of mortality relatively well in statistical analyses. Calf percentage and the yearly survival index were the means of two rein- deer management districts in three of the study areas. These reindeer management districts were in the Sodankylä area Oraniemi and Sattasniemi, in the Ivalo area Ivalo and Hammastunturi, and in the Kevo area Paistunturi and Kaldoaivi. In the Muonio study area only the Muonio reindeer management district was included in the analy- sis (Fig 1). Alternations of mean calf percentage, mortality index and reindeer density in the period 1962–1987 are given in Fig 2. On the basis of the data from the Finnish Metorological Institute, the mean temperature of the coldest month—February—during 1961– 1990 has been –13.6 °C in Sodankylä, –14.5 °C in Muonio, –13.0 °C in Ivalo and –14.1 °C in Kevo. The mean temperature of the warmest month— July—in the same period has been +14.1 °C in Sodankylä, +13.7 °C in Muonio, +13.7 °C in Ivalo and +12.7 °C in Kevo. Mean snow depth in March during 1961–1990 has been 72 cm in Sodankylä, 68 cm in Muonio, 65 cm in Ivalo and 66 cm in Kevo. The Sodankylä, Ivalo and Muonio study areas lie within the northern Boreal zone. Both in the Muonio and Ivalo study areas, dry and very dry pine forests as winter pastures are much more common than in the Sodankylä area, where sub- mesic and mesic spruce forests have replaced these pastures in late winter. The Kevo area belongs to the Fjell Lapland birch forest zone where dry and very dry mountain birch for- ests (Betula pubescens var tortuosa) and heaths above the tree line form the main winter pastures. During the study period reindeer grazing on nat- ural pastures in the Sodankylä area obtained their food in early and mid-winter mainly by digging reindeer lichens (Cladonia spp.) and other terres- tial plants (like grass, sedges and dwarf shrubs), but in late winter they browsed mainly arbo- real lichens (Alectoria, Bryoria spp.) in spruce forests. In the other three study areas, reindeer lichens and other terrestial plants dominated the diet of reindeer during the whole winter (Helle & Saastamoinen 1979; Kojola et al. 1995). Very large open mires are typical of the Sodankylä area and form the most important summer pasture type there. Although there are considerably fewer mires in the Muonio area than in the Sodankylä area, mires still have some importance as summer pastures in Muonio. In contrast, in the Ivalo area there are very few mires areas and submesic forests replace mires as summer pastures. In the Kevo area reindeer use mainly submesic birch forests and some mires as summer pastures. The terrain in the Muonio area and even more so in the Ivalo and Kevo areas is much more rugged and hilly than in the Sodankylä area, which is relatively fl at. Dependency of calf percent and the yearly sur- vival index on weather and snow parameters in the previous year were fi rst analysed with simple regression models. Then reindeer density and only those parameters correlating signifi cant- ly (P ≤ 0.05) with the calf percentage and sur- vival index were used in GLM analyses. After that, GLM analyses with backwards stepwise (with alpha-to-remove = 0.05) were made with the same independent variables in order to fi nd the most predictive parameters for reproduction and mortality. Results Contrary to what could be expected, high reindeer densities did not seem to correlate negatively with a low calf percentage in any area (Table 1). In the Sodankylä and Ivalo areas, calf percentage was high when reindeer density was high. When look- ing only at the effects of weather and snow con- ditions on calf percentage in the stepwise GLM analyses, the higher the number of very warm days or mean temperatures in summer, the lower 228 Effects of weather and snow conditions on reproduction and survival of reindeer Table 1. Effect of reindeer density and weather on calf percentage (calves / 100 females) in the study areas, 1962–1987: results of ordinary GLM analyses and GLM analyses with stepwise command. RDEN stands for reindeer density (reindeer / km²) during the previous winter. MTJULY is the mean temperature (°C) in July. DAY0DEC is the number of days in December when the mean temperature is above 0 °C. DIFTDECNOV is the difference in mean temperatures between December and November. MAX25JUL is the number of days in July when the maximum temperature is above 25 °C. SNACWIN is the sum of the monthly snow depths (on the 15th day of the month) from November to April (cm). MTJUNE is the mean temperature (°C) in June. MTAUG is the mean temperature (°C) in August. MAX25JUNAUG is the number of days from July to August when the maximum temperature is above 25 °C. DAY0MAY is the number of days in May when the mean temperature is above 5 °C. Study area Effect Coeffi cient SE Std. coef. Tolerance T P Sodankylä O rd in ar y G L M : Constant 45.367 20.623 0.000 – 2.200 0.040 RDEN 24.015 5.007 0.659 0.926 4.797 0.000 MTJULY –2.258 1.393 –0.250 0.732 –1.620 0.121 DAY0DEC –0.061 1.352 –0.008 0.590 –0.045 0.964 DIFTDECNOV –0.531 0.430 –0.220 0.554 –1.236 0.231 R² = 0.650, n = 25, F = 9.294, P = 0.000 St ep w is e in : Constant 60.929 17.604 0.000 – 3.461 0.002 RDEN 25.272 4.829 0.694 1.000 5.233 0.000 MTJULY –3.262 1.195 –0.362 1.000 –2.731 0.012 R² = 0.613, n = 25, F = 17.458, P = 0.000 Muonio O rd in ar y G L M : Constant 56.692 21.406 0.000 – 2.555 0.020 RDEN 6.626 3.813 0.326 0.759 1.738 0.099 MAX25JUL –1.321 1.121 –0.264 0.532 –1.178 0.254 DAY0DEC –1.157 3.287 –0.088 0.432 –0.352 0.729 DIFTDECNOV –0.093 0.514 –0.040 0.550 –0.180 0.859 SNACWIN –0.062 0.055 –0.281 0.432 –1.129 0.274 R² = 0.518, n = 24, F = 3.876, P = 0.015 St ep w is e in : Constant 78.725 11.163 0.000 – 7.052 0.000 MAX25JUL –2.031 0.840 –0.406 0.951 –2.418 0.025 SNACWIN –0.096 0.037 –0.439 0.951 –2.418 0.016 R² = 0.436, n = 24, F = 8.126, P = 0.002 Ivalo O rd in ar y G L M : Constant 79.392 28.931 0.000 – 2.744 0.013 RDEN 13.056 2.976 0.585 0.709 4.387 0.000 MTJUNE –0.200 1.275 –0.027 0.439 –0.157 0.877 MTAUG –2.465 1.865 –0.187 0.632 –1.321 0.203 MAX25JUNAUG –0.897 0.550 –0.287 0.406 –1.631 0.120 PRECMAY 0.129 0.143 0.124 0.672 0.903 0.378 SNACWIN –0.069 0.032 –0.265 0.828 –2.148 0.046 R² = 0.773, n = 25, F = 10.212, P =0.000 St ep w is e in : Constant 32.262 4.246 0.000 – 7.598 0.000 RDEN 15.502 2.689 0.695 0.982 5.766 0.000 MAX25JUNAUG –1.147 0.376 –0.367 0.982 -3.049 0.006 R² = 0.686, n = 25, F = 24.050, P = 0.000 Kevo O rd in ar y G L M : Constant 1.126 15.543 0.000 – 0.072 0.943 RDEN 5.278 3.354 0.294 0.810 1.574 0.131 DAY0DEC –3.037 1.691 –0.316 0.916 –1.796 0.088 DAY0MAY 1.322 0.639 0.377 0.854 2.070 0.052 R² = 0.433, n = 24, F = 5.083, P = 0.009 St ep w is e in : Constant 6.004 15.758 0.000 – 0.381 0.707 DAY0DEC –3.783 1.679 –0.394 0.995 –2.253 0.035 DAY0MAY 1.700 0.612 0.485 0.995 2.777 0.011 R² = 0.362, n = 24, F = 5.966, P = 0.009 229Kumpula 2003: Polar Research 22(2), 225–233 Fig. 2. (a) Variation in mean yearly calf percentage (calves/ 100 females) in round-ups (October–January); (b) the sur- vival index of reindeer; and (c) reindeer density (reindeer/km² land area, winter stock) between 1962–1987 in the study areas. Original data were obtained from the statistics of the Finnish Reindeer Herders’ Association. (a) (b) (c) 230 Effects of weather and snow conditions on reproduction and survival of reindeer was the calf percentage in the next year round- ups in all forested areas in Sodankylä, Muonio and Ivalo (Table 1). This did not seem to apply in the cooler and more mountainous Kevo area. In the Muonio area, the calf percentage was also reduced by winters with abundant snow accumu- lation. In the Kevo area, the greater the number of warm days (mean T > 0 °C) in December and the lower their number in May, the lower was the calf percentage in the next autumn round-ups. Of all weather variables, abundant snow accu- mulation during winter was the only one which seemed to reduce the survival index of reindeer in both the Sodankylä and Ivalo areas (Table 2). In the Sodankylä area, where reindeer used mainly arboreal lichens in late winter, abundant snow accumulation in March and April seemed to be the most important factor for high mortality. In the Ivalo area, where reindeer got their food by digging (cratering) throughout the winter, abun- dant snow accumulation during the whole winter period increased mortality. Also, high reindeer densities seemed to increase mortality in the Sodankylä area. In the Muonio area, mortality of reindeer was increased when the number of warm days in December was high. In the Kevo area, neither reindeer density nor any weather or snow parameters could explain yearly differences in the survival index. Discussion On the basis of this work and previous studies (e.g. Kumpula & Nieminen 1992; Helle & Kojola 1993), a marked increase in reindeer density in Finland since the mid-1970s did not seem to have any negative effect on the reproduction of rein- deer. This is probably associated with changes in management practice, including supplemen- tal feeding in the southern half of the area (Helle & Kojola 1993), and calf harvest (Kojola & Helle 1993; Kumpula et al. 1998), which made the herds strongly female-biased and thus improved the survival rate of the stock. It may also be relat- ed to the favourable climatic conditions prevail- ing during the 1980s (Helle et al. 2001). It is very probable that when there is a period of favoura- ble weather and snow conditions, as in our 1980s study period, reproduction is not directly affect- ed by reindeer density. It has been also demon- strated in Norway that the reproduction of adult female reindeer was not controlled by population density, although sub-adult fertility was reduced at high animal densities (Skogland 1985). In our study, survival of reindeer was reduced at high reindeer densities only in the Sodankylä area, where winter pastures were located both in pine and spruce forests. This density dependence may be linked to the fact that the amount of lichen pasture per reindeer in the Sodankylä area has been the smallest of the four study areas, which may have caused more competition between rein- deer in pastures at high densities. Due to the lack of lichen pastures, reindeer in the Sodankylä area have also had to graze mainly arboreal lichen in late winter. It may also be supposed that reduction of survival has probably been caused by increased juvenile mortality in winter since juvenile surviv- al has been observed to be the most sensitive to changes in reindeer density (Skogland 1985). Some clear dependence between local climat- ic and productivity factors of the semi-domesti- cated reindeer stock was found. In all forested areas, summer temperatures had a clear effect on calf production. High mean temperatures and large number of very warm days during summer months seemed to reduce calf production. Warm weather activates blood-sucking insects and keeps reindeer continuously moving in dense herds and may reduce their grazing and resting times (Helle & Tarvainen 1984; Helle & Kojola 1994; Mörschel & Klein 1997; Mörschel 1999). This may exhaust especially smaller calves if calf marking also occurs during a very warm period. Especially in woodland area where where high, open, windy ridges and hilltops are not availa- ble, blood-sucking insects and high temperatures in summer can form a very exhausting combina- tion for calves. Snow accumulation in the previous winter seemed to be an important factor for determining calf percentage or mortality in all three forested areas. The importance of snow depth in the pre- vious winter for ungulate reproduction and body weight has been identifi ed earlier (Helle & Kojola 1994; Adams & Dale 1998; Loison et al. 1999); deeper snow means usually more digging for food and thus more energy costs (Kumpula 2001b). However, it seems that in the Kevo area—where there are a lot of open, windy ridges and hilltops with shallow snow—snow accumulation itself has not had any direct effects on calf production or mortality. The Kevo Meteorological Station is located in a deep valley, which can mean that snow conditions in the adjacent mountain area are 231Kumpula 2003: Polar Research 22(2), 225–233 inaccurately registered. However, snow accumu- lation can affect calf production even in moun- tain areas: one recent study (Helle et al. 2001) has shown that snowy winters reduced the calf per- centage and affected yearly change in popula- tion size in the Käsivarsi mountain area, which is located nearer the western coast of northern Scandinavia than our study area. Generally, there seems to be three important periods for succesful reproduction and survival in a year: early winter, spring and mid-summer. Weather and snow conditions, particularly in December, essentially affect the development of those conditions which regulate the availability and quality of terrestrial winter food (Kumpula, Parikka et al. 2000). If there are warm Decem- bers with temperatures fl uctuating around 0 °C, a hard snow or ice layer can form on vegetation which then hampers foraging during the whole winter. This is probably a more important factor than snow accumulation itself in open, high pas- ture areas compared to woodland pasture areas. Especially in woodland areas, a thick snow layer on unfrozen soil in early winter can hamper dig- ging but may also promote the growth of cer- tain micro-fungi on vegetation (Kumpula, Parik- ka et al. 2000). These micro-fungi may then be harmful for the reindeer health. The body con- dition of reindeer females during calving is thus much affected by weather and snow conditions in late autumn and early winter. For reindeer herd- ers, early winter vegetation and snow conditions, which may then affect the availability and quali- ty of terrestrial food during the rest of the winter, Table 2. Effect of reindeer density and weather on the survival index of reindeer in the study areas, 1962–1987: results of ordinary GLM analyses and GLM analyses with stepwise command. RDEN stands for reindeer density (reindeer / km²) during the previous winter. SNACMAYAPR is the sum of the monthly snow depths (on 15th day of the month) from March to April (cm). DAY0DEC is the number of days in December when the mean temperature is above 0 °C. SNACWIN is the sum of the monthly snow depths (on the 15th day of the month) from November to April (cm). Study area Effect Coeffi cient SE Std. coef. Tolerance T P Sodankylä O rd in ar y G L M : Constant 110.962 20.150 0.000 . 5.507 0.000 RDEN –16.681 7.584 –0.322 0.986 –2.199 0.039 SNACMAYAPR –0.587 0.121 –0.712 0.986 –4.859 0.000 R² = 0.556, n = 24, F = 13.127, P = 0.000 St ep - w is e in : the same as above Muonio O rd in ar y G L M : Constant 0.471 20.604 0.000 . 0.023 0.982 RDEN 1.449 4.128 0.060 0.839 0.351 0.729 DAY0DEC –11.331 3.534 –0.692 0.522 –3.206 0.004 SNACWIN –0.003 0.059 –0.010 0.470 –0.046 0.964 St ep w is e in : R² = 0.514, n = 24, F = 7.053, P = 0.002 Constant 2.537 2.865 0.000 – 0.885 0.386 DAY0DEC –11.702 2.443 –0.714 1.000 –4.789 0.000 R² = 0.510, n = 24, F = 22.935, P = 0.000 Ivalo O rd in ar y G L M : Constant 88.799 26.188 0.000 – 3.391 0.003 RDEN –7.436 7.127 -0.188 0.996 –1.043 0.309 SNACWIN –0.262 0.086 -0.549 0.996 –3.056 0.006 R² = 0.324, n = 24, F = 5.032, P = 0.016 St ep w is e in : Constant 78.484 24.299 0.000 – 3.230 0.004 SNACWIN –0.257 0.086 –0.538 1.000 –2.990 0.007 R² = 0.289, n = 24, F = 8.940, P = 0.007 232 Effects of weather and snow conditions on reproduction and survival of reindeer have always been very important to estimate. In our data, the number of warm days in Decem- ber as well as the difference in average temper- atures between December and November proba- bly indicate something about the development of snow conditions as well as soil temperatures and the formation of permanent ground frost in soil in early winter. May is another important month for successful calf production especially in the northernmost part of the Finnish reindeer management area. After a warm, rainy May, the calf percentage seems to be high. This kind of weather probably causes rapid snow melt and the early emergence of green vegetation (Kumpula & Nieminen 1992) or can simply improve the availability of winter food when snow melts earlier (Helle et al 2001). This will improve the body condition of reindeer females and thus favour calving. This study indicates that permanent climate changes could affect woodland and open, moun- tainous areas in different ways. There should be broader knowledge of the effects of local weath- er conditions, and their variations, on vegeta- tion and reindeer populations and in which way the potential global climatic variation or change affects local weather and snow conditions. One index of global climate fl uctuations, the North Atlantic Oscillation (NAO), has been observed to correlate with plant phenology (Post & Stenseth 1998) and snow conditions (Mysterud et al. 2000) in Norway. Plants bloomed earlier following pos- itive NAO (warm, wet) winters, but the timing of fl owering varied with the NAO more strongly in southern than in northern Norway. Early-bloom- ing plants were more strongly infl uenced by the NAO than late-blooming plants. Snow depth was negatively correlated with the NAO at low alti- tudes but this relationship was reversed at high altitudes, giving a positive correlation between snow depth and the NAO. It has been also shown that positive NAO win- ters with high precipitation and warm weather are mostly adverse for ungulate populations in north- ern latitudes (excluding the coast of Norway and the Isle of Rum, Scotland), reducing body weight, reproduction and population growth rate and increasing mortality (Post & Stenseth 1999; Mys- terud et al. 2000). The NAO index has also been observed to have a strong impact on the popula- tion dynamics of semi-domesticated reindeer in the Käsivarsi area, northern Finland (Helle et al. 2001). Reduction of calf production and popula- tion size followed both highly negative and posi- tive NAO winters. This work has shown that variation in local weather and snow conditions, especially in early winter and spring, as well as snow accumula- tion during winter, could have a strong impact on the reindeer population in northern Finland. The clear impacts of summer temperatures on calf production of reindeer were also demonstrated. It is evident that if global climatic change means higher instability of weather in early winter and more snow during the whole winter in northern latitudes, conditions become more diffi cult for reindeer. These effects can be even more adverse if summer temperatures rise high. On the other hand, if climate change brings earlier snow melt- ing in spring, it can make some conditions more favourable for reindeer. Acknowledgements.—We wish to thank Jukka Siitari and Tuomo Haurunen from the Reindeer Research Station at the Finnish Game and Fisheries Research Institute for help in handling the data. For reindeer census data we thank the Finnish Reindeer Herders’ Association. Gordon Roberts is thanked for improving the English. This work is partly con- nected to the project no: 52759 in the SUNARE research pro- gramme, which is supported by the Finnish Academy. References Aanes, R., Sæther, B.-E. & Øritsland N. A. 2000: Fluctua- tions of an introduced population of Svalbard reindeer: the effects of density dependence and climatic variation. Ecog- raphy 23, 437–443. Adamczewski, J. Z., Gates, C. C., Soutar, B. M. & Hudson, R. J. 1988: Limiting effects of snow on seasonal habitat use and diets of caribou (Rangifer tarandus groenlandi- cus) on Coats Island, Northwest Territories, Canada. Can. J. Zool.66, 1986–1996. Adams L. G. & Dale B. W. 1998: Reproductive perform- ance of female Alaskan caribou. J. Wildl. Manage. 62, 1184–1195. Alpe, M. J., Kingery, J. M. & Mosley, J. C. 1999: Effects of summer sheep grazing on browse nutritive quality in autumn and winter. J. Wildl. Manage. 63, 346–354. Augustine, D. J., Frelich, L. E. & Jordan, P. A. 1998: Evi- dence for two alternate stable states in an ungulate grazing system. Ecol. Appl. 8, 1260–1269. Behnke, R. H. & Scoones, I. 1992: Rethinking range ecology: implications for rangeland management in Africa. Int. Inst. Environ. Dev. Pap. 33. Crête, M. & Doucet, G. J. 1998: Persistent suppression in dwarf birch after release from heavy summer browsing by caribou. Arct. Alp. Res. 30, 126–132. 233Kumpula 2003: Polar Research 22(2), 225–233 Ellis, J. E., Coughenour, M. B. & Swift, D. M. 1991: Cli- mate variability, ecosystem stability, and the implications for range and livestock development. In R.H. Behnke et al. (eds.): Range ecology at disequilibrium: new models of nat- ural variability and pastoral adaption in African savannas. Pp. 31–41. London: Overseas Development Institute. Fynn, R. W. S. & O’Connor, T. G. 2000: Effect of stocking rate and rainfall on rangeland dynamics and cattle per- formance in a semi-arid savanna, South Africa. J. Appl. Ecol. 37, 491–507. Helle, T. 1980: Laiduntilanteen muutokset ja riskinotto Suomen porotaloudessa. (Changes in the state of grazing areas and risk-taking in Finnish reindeer management. In Finnish with English summary.) Lapin tutkimusseuran vuosikirja XXI, 13–22. Helle, T. & Saastamoinen, L. 1979: The winter use of food resources of semi-domestic reindeer in northern Finland. Commun. Inst. For. Fenn. 95, 1–26. Helle, T. & Tarvainen, L. 1984: Effects of insect harassment on weight gain and survival of reindeer calves. Rangifer 4, 24–27. Helle, T. & Kojola, I. 1993: Reproduction and mortality of Finnish semi-domesticated reindeer in relation to density and management strategies. Arctic 46, 72–77. Helle, T. & Kojola, I. 1994: Body mass variation in semi- domesticated reindeer. Can. J. Zool. 72, 681–688. Helle, T., Kojola, I. & Timonen, M. 2001: Lumipeitteen vai- kutus Käsivarren porolukuihin: mikä on Pohjois-Atlantin säävaihtelun (NAO) merkitys? (Impact of snow cover on the reindeer population in Käsivarsi, NW Finland: is the North Atlantic Oscillation (NAO) involved? In Finnish with English summary.) Suom. Riista 47, 75–85. Henry, G. H. R. & Gunn, A. 1991: Recovery of tundra vegeta- tion after overgrazing by caribou in Arctic Canada. Arctic 44, 38–42. Klein, D. R. 1968: The introduction, increase, and crash of rein- deer on St. Matthew Island. J. Wildl. Manage. 32, 350–367. Kojola, I. & Helle, T. 1993: Calf harvest and reproductive rate of reindeer in Finland. J. Wildl. Manage. 57, 451–453. Kojola, I., Helle, T., Niskanen, M. & Aikio, P. 1995: Effects of lichen biomass on winter diet, body mass and reproduction of semi-domesticated reindeer Rangifer t. tarandus in Fin- land. Wildl. Biol. 1, 33–38. Kumpula, J. 2001a: Productivity of the semi-domesticated rein deer (Rangifer t. tarandus L.) stock and carrying cap- acity of pastures in Finland during 1960–1990’s. Acta Univer sitatis Ouluensis A 375, Scientiae Rerum Natural- ium. Ph.D. thesis, University of Oulu. Kumpula, J. 2001b: Winter grazing of reindeer in woodland lichen pasture: effect of lichen availability on the condition of reindeer. Small Ruminant Res. 39, 121–130. Kumpula, J., Colpaert, A. & Nieminen, M. 1998: Reproduc- tion and productivity of semi-domesticated reindeer in northern Finland. Can. J. Zool.76, 269–277. Kumpula, J., Colpaert, A. & Nieminen, M. 2000: Condition, potential recovery rate, and productivity of lichen (Clado- nia spp.) ranges in the Finnish reindeer management area. Arctic 53, 152–160. Kumpula, J., Colpaert, A. & Nieminen, M. 2002: Productivity factors of the Finnish semi-domesticated reindeer (Rangif- er t. tarandus) stock during the 1990s. Rangifer 22, 47–56. Kumpula, J. & Nieminen, M. 1992: Pastures calf production and carcass weights of reindeer calves in the Oraniemi co- operative, Finnish Lapland. Rangifer 12, 93–104. Kumpula, J., Parikka, P. & Nieminen, M. 2000: Occurence of certain microfungi on reindeer pastures in northern Fin- land during winter 1996–97. Rangifer 20, 3–8. Leader-Williams, N., Smith, R. I. L. & Rothery, P. 1987: Infl u- ence of introduced reindeer on the vegetation of South Georgia: results from a long-term exclusion experiment. J. Appl. Ecol. 24, 801–822. Loison, A., Langvatn, R. & Solberg, E. J. 1999: Body mass and winter mortality in red deer calves: disentangling sex and climate effects. Ecography 22, 20–30. Manseau, M., Huot, J. & Crête, M. 1996: Effects of summer grazing by caribou on composition and productivity of veg- etation: community and landscape level. J. Ecol. 84, 503– 513. Messier, F. 1991: On the concepts of population limitation and population regulation as applied to caribou demography. In C. E. Butler & S. P. Mahoney (eds.): Proceedings of the Fourth North American Caribou Workshop. Pp. 260–277. St. John’s: Newfoundland Labrador Wildlife Division. Milner, J. M., Elston, D. A. & Albon, S. D. 1999: Estimating the contributions of population density and climatic fl uctu- ations to interannual variation in survival of Soay sheep. J. Anim. Ecol. 68, 1235–1247. Moxnes, E., Danell, Ö., Gaare, E. & Kumpula, J. 2001: Opti- mal strategies for the use of reindeer rangelands. Ecol. Model. 145, 225–241. Mörschel, F. M. 1999: Use of climatic data to model the pres- ence of oestrid fl ies in caribou herds. J. Wildl. Manage. 63, 588–593. Mörschel, F. M. & Klein, D. R. 1997: Effects of weather and par a sitic insects on behaviour and group dynamic of car- ibou of the Delta Herd, Alaska. Can. J. Zool. 75, 1659– 1670. Murdock, W. W. 1994: Population regulation in theory and practice. Ecology 75, 271–287. Mysterud, A., Yoccoz, N. G., Stenseth, N. C. & Langvatn, R. 2000: Relationships between sex ratio, climate and density in red deer: the importance of spatial scale. J. Anim. Ecol. 69, 959–974. Ouellet, J.-P., Heard, D. C., Boutin, S. & Mulders R. 1997: A comparison of body condition and reproduction of caribou on two predator-free islands. Can. J. Zool. 75, 11–17. Portier, C., Festa-Bianchet, M., Gaillard, J.-M., Jorgenson, J. T. & Yoccoz, N. G. 1998: Effects of density and weather on survival of bighorn sheep lambs (Ovis canadensis). J. Zool. 245, 271–278. Post, E. & Stenseth, N. C. 1998: Large-scale climatic fl uctu- ation and population dynamics of moose and white-tailed deer. J. Anim. Ecol. 67, 537–543. Post, E. & Stenseth, N. C. 1999: Climatic variability, plant phe- n ol ogy, and northern ungulates. Ecology 80, 1322–1339. Reimers, E. 1982: Winter mortality and population trends of reindeer on Svalbard, Norway. Arct. Alp. Res. 14, 295–300. Sæther, B.-E. 1997: Environmental stochasticity and popu- lation dynamics of large herbivores: a search for mecha- nisms. Trends Evol. Ecol. 12, 143–149. Skogland, T 1985: The effects of density dependent resource limitations on demography of wild reindeer. J. Anim. Ecol. 54, 359–374. Skogland, T. 1986: Density dependent food limitation and max imal production in wild reindeer herds. J. Wildl. Manage. 50, 314–319. Turchin, P. 1995: Population regulation: old arguments and a new synthesis. In N. Cappuccino & P. W. Price (eds.): Population dynamics. Pp. 19–40. San Diego, CA: Aca- demic Press. << /ASCII85EncodePages false /AllowTransparency false /AutoPositionEPSFiles true /AutoRotatePages /All /Binding /Left /CalGrayProfile (Dot Gain 20%) /CalRGBProfile (sRGB IEC61966-2.1) /CalCMYKProfile (U.S. Web Coated \050SWOP\051 v2) /sRGBProfile (sRGB IEC61966-2.1) /CannotEmbedFontPolicy /Warning /CompatibilityLevel 1.4 /CompressObjects /Tags /CompressPages true /ConvertImagesToIndexed true /PassThroughJPEGImages true /CreateJDFFile false /CreateJobTicket false /DefaultRenderingIntent /Default /DetectBlends true /ColorConversionStrategy /LeaveColorUnchanged /DoThumbnails false /EmbedAllFonts true /EmbedJobOptions true /DSCReportingLevel 0 /SyntheticBoldness 1.00 /EmitDSCWarnings false /EndPage -1 /ImageMemory 1048576 /LockDistillerParams false /MaxSubsetPct 100 /Optimize true /OPM 1 /ParseDSCComments true /ParseDSCCommentsForDocInfo true /PreserveCopyPage true /PreserveEPSInfo true /PreserveHalftoneInfo false /PreserveOPIComments false /PreserveOverprintSettings true /StartPage 1 /SubsetFonts true /TransferFunctionInfo /Apply /UCRandBGInfo /Preserve /UsePrologue false /ColorSettingsFile () /AlwaysEmbed [ true ] /NeverEmbed [ true ] /AntiAliasColorImages false /DownsampleColorImages true /ColorImageDownsampleType /Bicubic /ColorImageResolution 300 /ColorImageDepth -1 /ColorImageDownsampleThreshold 1.50000 /EncodeColorImages true /ColorImageFilter /DCTEncode /AutoFilterColorImages true /ColorImageAutoFilterStrategy /JPEG /ColorACSImageDict << /QFactor 0.15 /HSamples [1 1 1 1] /VSamples [1 1 1 1] >> /ColorImageDict << /QFactor 0.15 /HSamples [1 1 1 1] /VSamples [1 1 1 1] >> /JPEG2000ColorACSImageDict << /TileWidth 256 /TileHeight 256 /Quality 30 >> /JPEG2000ColorImageDict << /TileWidth 256 /TileHeight 256 /Quality 30 >> /AntiAliasGrayImages false /DownsampleGrayImages true /GrayImageDownsampleType /Bicubic /GrayImageResolution 300 /GrayImageDepth -1 /GrayImageDownsampleThreshold 1.50000 /EncodeGrayImages true /GrayImageFilter /DCTEncode /AutoFilterGrayImages true /GrayImageAutoFilterStrategy /JPEG /GrayACSImageDict << /QFactor 0.15 /HSamples [1 1 1 1] /VSamples [1 1 1 1] >> /GrayImageDict << /QFactor 0.15 /HSamples [1 1 1 1] /VSamples [1 1 1 1] >> /JPEG2000GrayACSImageDict << /TileWidth 256 /TileHeight 256 /Quality 30 >> /JPEG2000GrayImageDict << /TileWidth 256 /TileHeight 256 /Quality 30 >> /AntiAliasMonoImages false /DownsampleMonoImages true /MonoImageDownsampleType /Bicubic /MonoImageResolution 1200 /MonoImageDepth -1 /MonoImageDownsampleThreshold 1.50000 /EncodeMonoImages true /MonoImageFilter /CCITTFaxEncode /MonoImageDict << /K -1 >> /AllowPSXObjects false /PDFX1aCheck false /PDFX3Check false /PDFXCompliantPDFOnly false /PDFXNoTrimBoxError true /PDFXTrimBoxToMediaBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXSetBleedBoxToMediaBox true /PDFXBleedBoxToTrimBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXOutputIntentProfile () /PDFXOutputCondition () /PDFXRegistryName (http://www.color.org) /PDFXTrapped /Unknown /Description << /FRA /ENU (Use these settings to create PDF documents with higher image resolution for improved printing quality. The PDF documents can be opened with Acrobat and Reader 5.0 and later.) /JPN /DEU /PTB /DAN /NLD /ESP /SUO /ITA /NOR /SVE >> >> setdistillerparams << /HWResolution [2400 2400] /PageSize [612.000 792.000] >> setpagedevice