The impact of microclimate and soil on the ecology and evolution of an arctic plant

The arctic and alpine regions are predicted to experience one of the highest rates of climate change, and the arctic vegetation is expected to be especially sensitive to such changes. Understanding the ecological and evolutionary responses of arctic plant species to changes in climate is therefore a key objective. Geothermal areas, where temperature gradients naturally occur over small spatial scales, and without many of the confounding environmental factors present in latitudinal and other gradient studies, provide a natural experimental setting to examine the impact of temperature on the response of arctic-alpine plants to increasing temperatures. To test the ecological and evolutionary response of the circumpolar alpine bistort (Bistorta vivipara) to temperature, we collected plant material and soil from areas with low, intermediate, and high soil temperatures and grew them in all combinations at three different temperatures. At higher experimental soil temperatures, sprouting was earlier, and plants had more leaves. Sprouting was earlier in soil originating from intermediate temperature and plants had more leaves when grown in soil originating from low temperatures. We did not find evidence of local adaptation or genetic variation in reaction norms among plants originating from areas with low, intermediate, and high soil temperature. Our findings suggest that the alpine bistort has a strong plastic response to warming, but that differences in soil temperature have not resulted in genetic differentiation. The lack of an observed evolutionary response may, for example, be due to the absence of temperature-mediated selection on B. vivipara, or high levels of gene flow balancing differences in selection. When placed within the context of other studies, we conclude that arctic-alpine plant species often show strong plastic responses to spring warming, while evidence of evolutionary responses varies among species.

2) The bulbils planted in soils originating from heated areas have a higher probability of 1 0 2 sprouting, sprout earlier, and grow larger due to higher availability of mineralized Geothermal landscapes, where magma gets close to the surface and heats soils and ground 1 1 7 water, can act as hotspots of local adaptation (O'Gorman et al. 2014). Notably, within these 1 1 8 landscapes the temperature can vary strongly at small spatial scales, with temperatures 1 1 9 differing by several or tens of degrees Celsius between locations separated by only a few  The alpine bistort (Bistorta vivipara) is a perennial herb that occurs in many parts of  We collected plants and soil from Hengladalir, a geothermally active valley located a few  To investigate the effect of temperature and soil on plant performance, we designed a planted with two bulbils. We distributed bulbils from each mother plant equally across 1 5 4 treatment combinations. As maternal investment in bulbils may affect growth, we recorded 1 5 5 bulbil weight prior to planting.

5 6
We experimentally heated the plant pots in water baths. The water baths were 1 5 7 uniformly heated using a combination of aquarium heaters (Easyheater 75W, AQUAEL, Poland) and water circulation pumps (Compact 300, EHEIM, Germany). In total there were 1 5 9 nine water baths, three at each temperature level: i) low temperature (13°C), ii) intermediate  To avoid microbial contamination, we sealed the plant pots, but filled the lower 2.5 1 6 3 cm with gravel and water-absorbing cloth to allow for drainage. To allow water to flow effects, the location of the plant pots was regularly randomized within the three water baths 1 6 6 belonging to the same temperature treatment.

6 7
To disentangle the effect of soil biota from that of the abiotic soil environment (e.g.   To test for the impact of soil temperature and origin of plants and soil on bulbil weight was included as a covariate. To account for variation among mother plants, we 1 9 0 included the random factor 'Mother plant'. As we planted two bulbils in each plant pot, we probability of sprouting, we used a binomial distribution with a logit link; for day of 1 9 3 sprouting, plant height and leaf size, we used a normal distribution with an identity link; and 1 9 4 for the number of leaves, we used a Poisson distribution with a log link. We a priori planned to test for local adaptation in those cases where we found an  including all two-and three-way interactions. As above, we included the random factors  Bulbils were more likely to sprout and sprouted earlier when grown at higher soil 2 1 4 temperatures ( Fig 1AB,  temperatures had more leaves when plants were measured 126 days after planting (Fig. 1CD,  Fig. 2A). Plants had a higher 2 1 9 number of leaves at 126 days after planting when grown in soil from a low temperature origin 2 2 0 ( Fig. 2B, Table 1). At day 159, the end of the experiment, leaf size was interactively affected was not affected by any of the experimental factors (Table 1).

5
Plants originating from areas with low, intermediate, and high soil temperature did not 2 2 6 differ in probability of sprouting, day of sprouting or growth traits (Table 1). We also did not probability, day of sprouting or plant growth (Table 1).
Bulbil sprouting was later in the sterile soil than in the field soil, and sprouting was 2 3 1 earlier with experimental soil heating (Fig. S2, Table S1). We further detected a weak 2 3 2 interactive effect between original bulbil temperature, experimental soil temperature and soil 2 3 3 sterilization, suggesting that bulbils of a high temperature origin sprouted later than expected 2 3 4 when growing in soil originating from an area with low temperature, but only when the soil 2 3 5 was sterilized. We used a multi-factorial experiment to investigate the impact of microclimate and soil on 2 3 8 the immediate and evolutionary response of the alpine bistort. As expected, higher 2 3 9 experimental soil temperature advanced bulbil sprouting and led to plants with more leaves.

4 0
In contrast to our prediction, sprouting was earliest in soil originating from locations with biota.

4 7
Plants sprouted earlier and had more leaves with higher experimental soil 2 4 8 1 9