Introduced Caribou and Reindeer in the Aleutian Archipelago of Alaska PDF Download

Author: Mark Antone Ricca
Publisher:
ISBN: 9781303794544
Category :
Languages : en
Pages :

Get Book Here

Book Description
Introductions of mammalian herbivores to remote islands without predators allow a natural experiment to ask: 1) how herbivore populations persist and respond numerically and spatially when primarily under bottom-up control, and 2) how do the above-mentioned dynamics mediate feedbacks between plant and soil processes that could trigger a shift to a different state in ecosystems that have not ostensibly coevolved with herbivory? I took advantage of early- to- mid 20th century introductions of caribou and reindeer (Rangifer tarandus) to islands in the central Aleutian archipelago of Alaska to ask these questions.I used the Riney-Caughley 4-stage model (i.e., increase, K overshoot, decline, K re-equilibration) as a framework to test predictions of irruptive population growth and spatial expansion of caribou introduced to Adak Island. I utilized a time series of spatially explicit counts conducted intermittently over a 54-year period, whereby population size increased from 23 animals at the time of introduction to approximately 2900 animals by the year 2012. However, the pattern of numerical growth did not tightly fit the model predicted shape, and the population had not undergone all 4 stages. Population dynamics were characterized by two distinct periods of irruptive growth separated by a long stable period related to pulses in hunter-harvest rates. An unexpected pattern resembling logistic population growth occurred at the end of the time series that was most likely due to an additive combination of high levels of unreported hunter-harvest and intensifying density-dependent feedbacks on caribou fecundity and survival. Spatial use patterns during the post-calving season strongly supported model predictions, whereby high-density core areas expanded outwardly as population size increased. During the calving season, however, animals displayed marked site fidelity across the full range of population densities despite the availability of other suitable habitats for calving. Lastly, a small number of caribou emigrated to the adjacent island of Kagalaska at the end of the time series, which represents a new dispersal front in accordance with model predictions. I then asked if temporal and spatial variation in Rangifer herbivory accelerated ecosystem processes in maritime tundra to drive a feedback between increased graminoid production and rapid nitrogen cycling, or decelerated processes by stimulating shrub domination and slower N cycling. I measured plant and soil properties across three islands in the archipelago representing a chronosequence of elapsed time post-Rangifer introduction with distinct population dynamics within the Riney-Caughley framework nested within each island. The island chronosequence comprised: 1) Atka Island with ~100 years elapsed post-introduction, where Rangifer have experienced minimal harvest pressure that has likely contributed to the completion of all 4-irruptive stages, 2) Adak Island with ~ 50 years elapsed and kept at low density for ~ 35 years by harvest before undergoing rapid exponential growth characteristic of model stage-2, and 3) Kagalaska Island, which was set as a pseudo-control given the recent detection of very small numbers of caribou that emigrated from Adak. I also measured Rangifer spatial use within islands (indexed by pellet group counts) to determine how ecosystem processes responded to spatial variation in herbivory. Graminoids increased while dwarf-shrubs, lichens, and mosses decreased with herbivory history along the island chronosequence. Slow-growing Cladonia lichens that are highly preferred winter forage were decimated on both Rangifer-occupied islands. In contrast, increasing Rangifer spatial use within islands was correlated linearly with reductions in graminoid and forb biomass and non-linearly with reductions in dwarf-shrub biomass. However vegetation shifts associated with Rangifer along temporal and spatial gradients did not cascade strongly to affect belowground ecosystem function, whereby rates of net N-mineralization, NH4+ pools, and soil [delta]15N declined markedly along the chronosequence. Collectively, a multivariate ordination revealed strongly divergent ecosystem processes between Atka (characterized by lower fertility, fewer forbs and lichens) and Kagalaska (higher fertility, more abundant forbs, dwarf-shrubs, and lichens) that likely reflect a decrease in carrying capacity on Atka as predicted by the Riney-Caughley model. In contrast, ecosystem processes on Adak were intermediate between those on Atka and Kagalaska, which may indicate more transient alterations to plants and soil related to the relatively recent stage-2 irruption in Rangifer population size. These results support the hypothesis that Rangifer population persistence on islands is facilitated by successful exploitation of graminoid biomass as winter forage after palatable lichens are decimated, yet long-term grazing appears to reduce overall soil fertility. Finally, I asked how Rangifer modified or created heterogeneity in plant and soil processes at small-spatial scales on Adak via two discrete mechanisms. First, topographic variation stemming from small hilltop `knobs' that drain into semi-closed `swales' creates heterogeneity in forage resources that attract caribou. Use by caribou should then enhance or retard topographic-related heterogeneity in plant and soil processes depending on whether herbivory drives an acceleration or deceleration of ecosystem processes. Second, I asked if hunter-killed caribou-carcasses contributed to highly localized hotspots of soil and foliar nitrogen cycling, which was germane given the dramatic increase in recent harvest rates. Using a design of paired knob-swale plots in unreplicated drainages with high, low, or nominal caribou use, I found that high caribou use reduced heterogeneity in plant community structure but had little influence on variation in belowground processes associated with topography. Specifically, high use knobs had fewer and more compacted dwarf-shrub mats, reduced Cladonia lichen cover, and increased cover of deciduous shrubs and unpalatable lichens compared to all other plot types. However, these changes were not accompanied by higher biomass of graminoids and forbs that would be predicted if intense herbivory had an accelerating effect on ecosystem processes. High caribou use did not influence existing topography-related differences in inorganic soil-N pools or soil [delta]15N. Furthermore, caribou carcasses contributed to markedly elevated inorganic-N pools and higher foliar-N content despite driving greater net N-immobilization in comparison to control plots. Hence, large numbers of caribou carcasses scattered across the island landscape should add highly localized heterogeneity to N-cycling.