Shrub expansion in SW Greenland: the role of climate, grazing and other important drivers

Publikation: Bog/antologi/afhandling/rapportPh.d.-afhandlingForskning

Arctic regions have experienced higher temperatures in recent decades, and the warming trend is projected to continue in the coming years. Arctic ecosystems are considered to be particularly vulnerable to climate change. Expansion of shrubs has been observed widely in tundra areas across
the Arctic, and has a range of ecosystem effects where it occurs. Shrub expansion has to a large extend been attributed to increasing temperatures over the past century, while grazing and human disturbance have received less attention. Alnus viridis ssp. crispa is a common arctic species that
contributes to increasing shrub cover. Despite this, there is only limited experimental evidence that growth of the species responds to warming.
Plant populations in fragmented and isolated locations could face problems adapting to a warming climate due to limited genetic variation and restricted migration from southern locations. A. viridis represents an interesting case to study these effects. SW Greenland is a subarctic to low-arctic region with only limited increases in temperatures during the past decades, and observed climate trends being largely dependent on the observation period. In this region there is limited information on changes in shrub cover, and a long and complex land use history has to be taken into account.
In this thesis a combination of descriptive and experimental approaches are used to investigate:
I. Whether landscape scale shrub expansion occurred in SW Greenland, and to what extend it is influenced by factors like grazing and human disturbance;
II. which climatic factors control shrub growth in SW Greenland and whether these have improved sufficiently over the past century to allow shrub expansion;
III. whether growth of A. viridis is promoted by experimental warming;
IV. and whether plant genotypes from more southerly habitats are better adapted to climatic conditions in a warmer Greenland compared with local provenances.
To answer the first question historical photographs of vegetation in SW Greenland (1898–1974) were compiled. The photos were repeated in 2010 and 2011 and 64 photo pairs were cropped into 133 smaller units and classified by aspect, substrate stability, muskoxen grazing and human disturbance. The photo material was evaluated by 22 experts with respect to changes in shrub cover. The results revealed a general shrub cover increase in the whole dataset, but also in a subset including only undisturbed sites. Shrub cover increased most on E and SE facing slopes, in sites with stable substrate, in areas characterised by human disturbance and in areas without
muskoxen grazing. Aspect and human disturbances had the strongest effect on shrub expansion, followed by muskoxen and substrate type.
To answer second question, radial stem growth of three common arctic shrub species (A. viridis, Salix glauca, Betula pubescens) were examined. The samples were collected along a S–N gradient of ca. 1000 km in W Greenland. The observed growth fluctuations of the three species exhibited similar patterns over time indicating a common response of shrub vegetation to climate variation in W Greenland. The radial growth of A. viridis and S. glauca was analysed using standard dendrochronological methods (‘response function’) to identify the main climatic determinants of growth, including monthly mean temperature and precipitation. The influence of snow cover was investigated and negative effects were found. Hard wind events had no significant effects on growth of the two species. Significant climatic traits were summer temperatures but also certain precipitation variables depending on species. These climatic variables were used to model radial growth and to estimate the expected growth of A. viridis and S. glauca in W Greenland during 1890–2010. Overall, the estimated growth of the two species did not improve over the past 120 years, but considerable variation in growth conditions (including a recent 10–15 year period of favourable conditions) are conspicuous and match known historical patterns of the North Atlantic
Multidecadal Oscillation. Reasons for the observed shrub expansion in SW Greenland are discussed, and it is unlikely that the recent 10–15-year period of favourable conditions could be the most important factor responsible. It is likely that shrub expansion in SW Greenland is mainly caused by land use change, dating back from cessation of Norse settlers activities and extinction of caribou around 1900, albeit depending on regions, and possibly due to cessation of firewood collection. A delayed reaction to the ending of the little ice age cannot be excluded, but seems rather unlikely considering other studies from Greenland. Effects of global warming in SW Greenland must be studied over even longer time periods than the 120 years of the current study.
To answer the third question a novel method for passive warming was applied by using transparent bags installed on selected twigs. The experiment was conducted in SW Greenland in summer 2010 and 2011. In 2010 closed bags were used, while in the second season perforated bags were chosen to reduce the magnitude of passive warming. Shrubs on S vs. N facing slopes were investigated, including effects on terminal vs. side twigs. Radial growth of the annual rings of the twigs was measured, while growth during the experimental season divided by the growth during the previous season was used in the statistical analysis. Summer 2010 was the warmest measured in SW Greenland; although slightly cooler, summer 2011 was still the 7th warmest registered. Both summers were characterised by low precipitation. In 2010 the radial growth responded negatively to experimental warming due to deleterious temperatures caused by the treatment. The modified treatment (with perforated bags) in 2011 showed no negative effects, but did not result in higher growth either. There were no differences in the effect of warming comparing shrubs on S vs. N facing slopes or between terminal vs. side twigs. Due to higher summer temperatures, the growth in the control twigs was considerably higher in 2010 compa red to 2009. Although summer 2011 was slightly cooler, the growth remained on a similar level on S facing slopes but was slightly lower on the N facing slopes. It is concluded that radial growth of the shrubs was not limited by temperature in 2010 and possibly not in 2011. The limited effect of experimental warming was most likely caused by unusually high temperatures in the two growth seasons. The only partial heating of the shrubs could be responsible for the absence of effect of warming in 2011 on N facing slopes.
To answer the fourth question a transplantation experiment in SW Greenland was conducted. Three local provenances and three provenances collected ca. 1,400 km south in NE Canada were tested. The reaction to experimental warming in open top chambers (OTC) was investigated. Survival rates were reduced for plants exposed to OTC, but the devices had no significant effects on the growth and the effects of differed not among provenances. This suggests that during warm summers, plant establishment is not limited by temperatures, but rather by other factors like e.g. water availability. The sou thern provenances had higher growth under optimal greenhouse conditions prior to transplantation, but also after transplanting. This indicates that the southern provenances are better adapted to warm summer conditions in SW Greenland. Plant populations in Greenland could be restricted by local adaptation and low genetic variability which could limit adaptations to future climates.
ForlagDepartment of Plant and Environmental Sciences, Faculty of Science, University of Copenhagen
Antal sider135
StatusUdgivet - 2012

ID: 44070764