Response of plant species richness and primary productivity in shrublands along a north–south gradient in Europe to seven years of experimental warming and drought: reductions in primary productivity in the heat and drought year of 2003

We used a nonintrusive field experiment carried out at six sites – Wales (UK), Denmark (DK), the Netherlands (NL), Hungary (HU), Sardinia (Italy – IT), and Catalonia (Spain – SP) – along a climatic and latitudinal gradient to examine the response of plant species richness and primary productivity to warming and drought in shrubland ecosystems. The warming treatment raised the plot daily temperature by ca. 1 °C, while the drought treatment led to a reduction in soil moisture at the peak of the growing season that ranged from 26% at the SP site to 82% in the NL site. During the 7 years the experiment lasted (1999–2005), we used the pin‐point method to measure the species composition of plant communities and plant biomass, litterfall, and shoot growth of the dominant plant species at each site. A significantly lower increase in the number of species pin‐pointed per transect was found in the drought plots at the SP site, where the plant community was still in a process of recovering from a forest fire in 1994. No changes in species richness were found at the other sites, which were at a more mature and stable state of succession and, thus less liable to recruitment of new species. The relationship between annual biomass accumulation and temperature of the growing season was positive at the coldest site and negative at the warmest site. The warming treatment tended to increase the aboveground net primary productivity (ANPP) at the northern sites. The relationship between annual biomass accumulation and soil moisture during the growing season was not significant at the wettest sites, but was positive at the driest sites. The drought treatment tended to reduce the ANPP in the NL, HU, IT, and SP sites. The responses to warming were very strongly related to the Gaussen aridity index (stronger responses the lower the aridity), whereas the responses to drought were not. Changes in the annual aboveground biomass accumulation, litterfall, and, thus, the ANPP, mirrored the interannual variation in climate conditions: the most outstanding change was a decrease in biomass accumulation and an increase in litterfall at most sites during the abnormally hot year of 2003. Species richness also tended to decrease in 2003 at all sites except the cold and wet UK site. Species‐specific responses to warming were found in shoot growth: at the SP site, Globularia alypum was not affected, while the other dominant species, Erica multiflora, grew 30% more; at the UK site, Calluna vulgaris tended to grow more in the warming plots, while Empetrum nigrum tended to grow less. Drought treatment decreased plant growth in several studied species, although there were some species such as Pinus halepensis at the SP site or C. vulgaris at the UK site that were not affected. The magnitude of responses to warming and drought thus depended greatly on the differences between sites, years, and species and these multiple plant responses may be expected to have consequences at ecosystem and community level. Decreases in biodiversity and the increase in E. multiflora growth at the SP site as a response to warming challenge the assumption that sensitivity to warming may be less well developed at more southerly latitudes; likewise, the fact that one of the studied shrublands presented negative ANPP as a response to the 2003 heat wave also challenges the hypothesis that future climate warming will lead to an enhancement of plant growth and carbon sequestration in temperate ecosystems. Extreme events may thus change the general trend of increased productivity in response to warming in the colder sites.