Skip to main content

Vulnerability assessment of shrubland ecosystems in europe under climatic changes (VULCAN)


Fauna sampling in the six sites was carried out for 5-6 days per site in the period April 29 to July 23 2003. The sampling time for each site was chosen to correspond to late spring or early summer and to coincide with the beginning of the drought treatment. In each plot of the VULCAN sites 5 subplots (1260cm2) were selected. Composition, cover, age and height of the vegetation were used as basis for choosing the position of subplots emphasizing similarity between plots. A specially designed non-destructive suction method was used to sample the fauna on plants and soil surface. Within each subplot leaves and stems of all plants and an area of the soil/litter surfacewere systematically sucked by a mouthpiece connected to a fauna-trap and a vacuum-cleaner. Animals dropping from the plants during suction were sampled in plastic boxes placed under the plants before suction. A 10 cm deep soil sample (surface area 25cm2) was collected within each subplot and the soil fauna extracted by a standard high-gradient extraction method at the Mols Laboratory. Additional samplings were carried out in similar vegetation outside the experimental plots for comparison of the suction sampling method with conventional methods. The large material was sorted and the animals were counted to the level of larger taxonomic groups (classes, orders, families) and for selected taxa (Collembola, oribatid mites) to species level. Biomass was as a first approximation calculated by employing average values borrowed from the literature of individual weights representative for broad taxonomic groups. A more precise and site-specific measure of average individual weight representative of important fauna groups is being worked out based on microscopic measurements of body length and width. Comparisons of faunal abundance show several examples of significant differences between sites and between treatments within sites for broad fauna categories such as total invertebrates, total insects, plant feeders and detritus/microbial feeders (mainly soil fauna) as well as whole fauna groups such as Acari (mites), Collembola (springtails), leaf hoppers (cicadas), plant bugs, aphids, wasps, ants and spiders. Whether the responses of population density to manipulated drought and warming were negative, positive or non-significant depended on taxonomic group and site. The statistically significant comparisons showed that most individual taxonomic groups responded negatively in all sites to both types of climatic manipulations whereas opposite reactions in different sites were demonstrated for the composite assemblies of taxa: total insects, total fauna, plant feeders and detritus/microbial feeders. The cover percentage of individual plant species within subplots interacted with the treatments so that in some cases opposite effects on fauna populations were observed depending on vegetation type. Detailed analysis of the microarthropod taxa Acari and Collembola further demonstrated that different species may respond differently to the manipulations resulting in altered species composition and diversity due to the treatments. On the other hand, examples of opposite reactions of the same species in different sites were also observed. Biodiversity of Collembola (number of species) was lower in the drought treatment than in the control in the British, Dutch, Hungarian and Spanish site, but higher in the Danish and Italian sites. The species number in the warming treatment was equal to the control in the British and the Dutch sites, lower in the Hungarian and Spanish sites and higher in the Danish and Italian sites. The distribution between sites and treatments of crude biomass estimates based on average individual weights borrowed from the literature differed in composite taxa considerably from distributions based on abundance values because of a large variety in size found between the fauna groups composing these taxa. Less crude biomass estimates based on measurement of the actual size distribution within the fauna groups is still in progress. It can be concluded that the arthropod fauna as a whole and several taxonomic and trophic fauna groups as well as a number of microarthropod species responded significantly to both the warming- and the drought treatment but that the response pattern is complex depending on site and vegetation quality within sites. The observed responses of the fauna cannot immediately lead to conclusions about the effect on plant growth and survival. No examples of ongoing pest attack were observed at the sampling occasions in 2003 whereas consequences for the Calluna vegetation of earlier attacks of heather beetles were recorded in the Danish and Dutch site.5 mauscripts are planned for international scientific journals. A current PhD-project. The effect of stress on biodiversity (S. Bahrndorff 2004-2006) is based on the VULCAN fauna studies and includes plans for resampling in the Danish and Spanish sites.
Nitrate pollution of groundwater is an important environment issue in soil water chemistry in many north European countries. Temporarily increased nitrate concentration as a result of climatic change is therefore of socio-economic relevance. In the southern sites soil water drainage of pollutants is of minor environmental importance. Nevertheless, effort will be carried out to characterize soil water quality at these sites and to investigate any potential effect of the treatments on water chemistry.
A range of soil measurements have been made at all sites to both quantify the services shrubland soils provide in terms of nutrient supply, carbon sink and water regulation, and the potential effects of future climate change. The results have provided a unique dataset of this understudied ecosystem type, which can be used for future parameterisation of models. In addition, the results from the climate change experimental treatments indicate that both warming and repeated summer drought will affect these services. Specifically, we have demonstrated the greater sensitivity of soils in northern regions to warming. Sensitivity of soil carbon loss (soil respiration) was greater by a factor of 6 at the Welsh site compared to sites in Spain and Italy. This was confirmed in the experimental treatments where a 1-degree temperature rise resulted in accelerated carbon loss in Wales and Denmark whereas a decline was observed in all other sites. Soil moisture thresholds were identified for all sites and were found to vary with water limitation of respiration at 5.5% to 25% g/g wet soil whilst excess water supressed soil respiration above 55% g/g wetsoil. Repeated summer drought generally decreased soil carbon loss except in Wales where stimulation in soil carbon loss was observed. Unexpectedly, there was also evidence of persistent reduced soil moisture in drought plots in both Wales and the Netherlands throughout the year suggesting either increased plant water use or a structural change in soil, which reduces soil moisture holding capacity. Thus, in some northern soils this creates a longterm positive feedback with enhanced soil CO2 fluxes throughout the whole year beyond the period of the summer drought. Net change in soil carbon storage also depends on the quantity and and fate of photosynthetically fixed carbon from plants. The effect of climate change on the quantity of litter is described in the plant work package. In the soils work package litter bag studies were used to identify the quantity of recently fixed carbon passed into long term stable soil organic matter pool. Warming increased the initial rate of litter decomposition, but did not affect the amount incorporated into the stable soil organic matter pool (known as the critical limit). We found plant species (i.e. litter quality) was the primary factor determining this critical limit thus shifts in overall plant productivity and/or a shift in species composition will be the key factors determining the input of crabon into the stable organic pool either offsetting or enhancing changes in carbon losses through soil respiration. Evidence from the Danish site suggests for northern sites there is a net reduction in the soil carbon sink as there has been a 15% drop in soil organic matter after 6 years of climate treatments. In most systems, soils are important sources of plant nutrients. We quantified the controls on the release of nitrogen from the soil store and also identified the impact of climate change treatments. Measurable positive net nitrogen production (i.e. mineralisation) was only found when soil moisture was within above 20% g / g wet soil or below 60% g /g wet soil. Warming will have greatest effect between these two threshold. Increased summer drought is therefore likely to have greatest impact on nitrogen availability for plants in our wettest site in Wales where soil moisture content is usually above the 60% water excess threshold. We found no evidence of changes in phosphorus demand by plants in response to climate treatments. The implications of these results are that northern sites are at greatest risk from both summer drought and warming with respect to all three services studied; carbon store, nutrient supply and water regulation emphasising the need for reductions in greenhouse gas emissions if these systems are to be protected. Further work is needed to combine the data from the plant work package to assess the overall ecosystem carbon balance for the different climate treatments. Treatments are planned to continue at all sites using national or other sources of funding. These results have been disseminated through attendance at a range of meetings, the VULCAN newsletters and a synthesis paper in the journal Ecosystems. A further paper is planned for submission in July 2005. A journal paper is also planned for the next VULCAN special issue.
We developed a rule-based system for developing and implementing socio-economic scenarios of land use change that could be applied to any region of Europe and developed a strategy for implementing this system to yield quantitative and spatial scenarios of land use change under various socio-economic futures (the GATE-UPI scenarios). We applied the socio-economic scenarios to the regions in which risk assessment was required: UK (Wales), Spain (Catalonia) and Hungary (Dune-Titse interfluve). We applied the full quantitative and spatial implementation to Wales, mapping the results of 61 socio-economic scenarios of land use change to 2050, and measuring their quantitative implications for heathland extent. With regard Spain, we mapped a narrower range of socio-economic scenarios in order to concentrate on the more complex dynamics associated with fire and its potential increase and effects on shrublands under climate change. In Hungary, the required spatial data did not exist, hence the socio-economic scenarios were applied in a non-spatial and more qualitative manner, albeit with quantification where possible. The results of these land use change scenarios then fed into the risk assessment methodology for each country. This combined the land use change results with - The results of the VULCAN experiments and - Results of studies into other impacts such as nitrogen deposition and fire in order to circumscribe all potential ecological pathways on the shrubland remaining under various land use change scenarios. This method allowed that shrubland might indeed increase under both land use and climate changes. The final stage in the risk assessment for all three countries involved predicting the more and less likely land use and ecological pathways. While this procedure involves considerable uncertainty, all underlying assumptions were made fully explicit in order to provide decision makers with high quality information and assessment of the future threats to European shrublands.
Spectral reflectance in waveband 400-950 nm was measured from nadir at height of 1m above the canopy with narrow angle (12 degrees) receiver during sunny days around midday. For spatial averaging receiver was scanned over the plot and at least 90 reflectance spectra per plot was recorded. A calibrated grey standard was used as a reference surface. Vegetation indices (NDVI, PRI) and their statistics were calculated. In order to get information about spectral properties of canopy background litter layer and bare soil outside the plots were measured. Measurements were made with clear weather between 10 AM and 3 PM. Reflectance measurements showed that differences in reflectance indices between treatments were considerably smaller than differences between sites. In general, higher NDVI as an indicator of biomass was accompanied with higher efficiencies of photosynthetic machinery measured with both, reflectance and fluorescence methods. Photochemical efficiency index PRI as well as NDVI differed significantly between heat and drought treatments in all sites showing that realised photon efficiency of photosynthetic machinery was higher under the heat treatment than under the draught treatment. Comparison of NDVI values with green biomass data obtained from pin-point measurements revealed expected relationship between these two parameters with strong saturation at high biomass conditions. Consequently, change in NDVI values can be interpreted as changes in green biomass with some limitations at high biomass conditions as in Wales where second order effects (e.g. intensive flowering of Calluna) may overshadow small changes in biomass. Gross sites comparison showed that the general effect of warming treatment is positive on both NDVI and green biomass whereas drought treatment had negative effect. Due to variability pattern and better averaging ability of reflectance measurements if compared to pin-pointing the treatment effect was statistically significant only in case of NDVI showing that in general, the treatment effects were tiny and can be revealed only with methods that allow better spatial averaging. The study showed that PRI had no tendency to saturate at high biomass conditions as NDVI making PRI more valuable tool for assessing vegetation status in such conditions. However, if to exclude data from Wales then the treatment effects on values of NDVI and on PRI are well correlated with each other showing that the treatment effect on the amount of green biomass is the most likely a result of changed photochemical efficiency of photosynthetic apparatus. Our study revealed great potential of reflectance measurements in monitoring climate change related alterations in European shrublands. Reflectance is easier to measure and has better averaging ability than direct measurements of biomass or physiological status of vegetation.
Biomass production: The experimental warming of approximately 1ºC induced up to 100% increase in the aboveground biomass growth in the northern sites, especially in the UK and the Netherlands site. Direct and indirect effects of warming such as longer growth season (leaf unfolding was advanced on average 15 days in all sites) and increased nutrient availability were likely to be particularly important in these northern sites, which tend to be temperature-limited. No effect of drought was found in the northern sites, which are not water-limited. Instead, there was no increase in total aboveground plant biomass growth in the water-stressed southern sites, as expected since warming increases water loss and temperatures are already close to the optimum for photosynthesis plant productivity in those ecosystems. The southern sites presented instead the most negative response to the drought treatment. However, it was also noticeable that a wet year such as 2002 allowed the drought plots to recover even better than the control plots. Regarding changes in plant tissue chemistry, the most outstanding result was a decrease in the tissue P concentrations with both warming and drought. The magnitude of the plant biomass response to the treatments was thus very sensitive to differences among sites (the cold-wet northern sites more sensitive to warming and the warm-dry southern sites more sensitive to drought), seasons (plant processes more sensitive to winter warming than during summer) and species. Roots: The effect of the treatments on root length density was not found in the soil cores of 2002 containing old roots, but was observed in the in-growth cores of 2003. The RLV of the new grown roots decreased significantly. This was most clear in the upper layer but in the lower layer this trend can also be seen. The warming treatment showed the same trend: in 2003 a decreasing tendency in most countries. The treatments apparently affect the RLV, which means that the capability of exploring the soil for nutrients and water, is affected. Drought probably negatively affects this capacity through the dying-off of fine roots. For warming, the mechanism may be different: increased organic matter decomposition may improve the availability of nutrients and subsequently reduce root growth. Community composition changes: There were different responses in phenological advances and in growth depending on species. These different responses have changed the competitive ability of species with likely community and ecosystem level consequences, some of which have already been found during this VULCAN project. For example, a decrease in seedling diversity in the warming and especially in the drought treatment was observed in a study of the seedling recruitments in the Spanish site. In the Welsh site there was an progressive loss of dominance of Empetrum nigrum which was negatively affected by winter warming whereas Calluna increased even more its dominance. In Denmark there was an interaction of warming and drought with herbivore attack which was enhanced in both treatments either because enhanced insect metabolism in warming or because enhanced larger consumption in response to decreasing nutritional value and biomass amount in drought. These changes induce a shift in the composition of the plant community towards a decrease in biodiversity. The community structural changes will finally affect the ecosystem functioning because they affect plant growth rates and therefore C uptake or litter decomposition and therefore nutrient availability. Decrease plant coverage and dry live fuel: erosion and fire risk: The plant coverage decreased in the drought plots in the southern sites at Spain and Italy thus increasing soil erosion risk through greater area of bare soil. The decrease in plant productivity, reproduction and recruitment means a reduced input of organic matter to the soil, and reduced soil plant coverage, leading to a decrease of the water retention capacity of the soil and to an increase of erosion risk. This is a serious risk for some areas of the Mediterranean region (see map). If the soil water content becomes less, there is a decrease in the productivity of the vegetation, further decreasing the organic matter input into the soil. Also related to erosion risk, there is the fire risk which was also increased under drought since there were decreases in water input and decreases in soil water content, increases in the ratio between dead and alive tissues, and decreases in the tissue water content making more flammable the plant biomass or ¿live fuel¿. Although there were no significant changes in the VOC content under the drought treatment, the increased VOC emissions after warming may also enhance the fire risk. In fact, in the last decades there has been a greater recurrence of fires partly linked to these effects (see satellite map). Wildfires and erosion are the current most important threads for these shrublands.
Warming treatment. The warming treatment was applied to 3 study plots of 20 m2 (5*4 meter) at each site. The warming plots were covered by a light scaffolding carrying a curtain reflecting the infrared radiation. The scaffolding was a frame of galvanised steel tubes covered by a polyethylene plastic tube to avoid leaching of contaminants from the frame into the plots. The curtain material consisted of 5mm wide aluminium strips knitted into a high density polyethylene (HDPE) mesh (ILS ALU, AB Ludvig Svensson, Kinna, Sweden). The curtains reflected 97% of the direct and 96% of the diffuse radiation and allowed transfer of water vapour. The curtains were coiled on a beam and connected to a motor. The motor was activated automatically by an electronic controller according to the following preset climatic conditions during the whole year: - Light intensity at sunset (light intensity < 0.4W m-2) the curtains were automatically drawn over the vegetation to reduce the loss of IR radiation. At sunrise the curtains were retracted to leave the plots open during the day. - Rain in order to keep the hydrological conditions in the plots unaffected a tipping bucket rain sensor activated the removal of the curtains in case of rain during the night (sensitivity <0.3 mm). When the rain stopped the curtains were automatically drawn over the vegetation again. - Wind to avoid damage to the curtains, a wind sensor activated the removal of the curtains when wind speeds exceeded 10m s-1 during the night. When the wind speed dropped below 10 m s-1 the curtains were automatically drawn over the vegetation again. The curtains were operated sequentially causing a delay time of c. 4 minutes from the first to the last curtain in the sequence. The height of the curtains matched the height of the vegetation at each site and was 0.6-1.0 meter above the soil surface. The study plots were open at all sides. All curtains were operated on 24V DC supplied by main power (Clocaenog, UK) or by solar panels (Mols, DK; Oldebroek, NL; Garraf, SP;). At Garraf and Oldebroek the warming treatments were stopped once or twice for 2-4 weeks during the dormant season for calibration of temperature sensors. Drought treatment. The drought treatment was applied to 3 study plots of 20m2 at each site for extended periods in the growing season. The drought plots were constructed similar to the warming treatments except that the curtain material was a transparent polyethylene (PE) plastic. The curtains were controlled by a rain sensor activating the curtains to cover the plots whenever it rained and remove the curtains when the rain stopped. The water collected by the curtains was removed from the area by gutters. The curtains were removed automatically if the wind speed exceeded 10m s-1. Beyond the time of the drought treatment the drought plots were run parallel to the control plots. Control. Parallel to the warming and drought treatments 3 untreated control plots were operated for comparison. The control plots were covered by a similar light scaffolding as for the warming and drought treatments, but with no curtain.
Objectives: - Involve end-users and decision makers in guiding the project and relating the results to end user experience and practices, - Produce guidelines for shrubland management, - Transfer results to potential end users. Involvement of end-user group (EUG) The EUG has caused the project to think broader, to focus on dissemination and the relevancy of the science and the project results to society Their first message was that guidelines for shrubland exist so the work should focus on transfer of knowledge to potential shrubland end users, land managers, nature conservationists, and decision makers on the implications of climate change. At the final workshop in December 2004, scientists, end users and invited guests contrasted the scientific results from the climate change experiments with other threats to shrublands to evaluate if the results call for new management actions. The results show that there are different effects of climate change on different shrubland ecosystems. Climate change influences i.e. the vegetation composition, productivity, phenology, fauna composition and soil processes. These changes may decrease shrubland ecosystem stability and content and may result in environmental problems. We observed a synergy between for example pollution, changes in land use and climatic changes so we cannot find attributes, which are unique for climatic changes. But climate change has increased the pressure on these vulnerable ecosystems. As a general advice the managers should listen to early warnings and take measures before the habitats have already been destroyed. To counteract the negative effects of climate change (the increasing pressure on shrublands) it is necessary to enhance the robustness of shrubland ecosystems through target oriented and prioritise management. One tool is linking fragmented heathland sites (increase connectivity) e.g. by reverting agricultural land and forest to heathland. To accomplish this there is a need to define and develop indicators that tell where change is happening, what the change is and what management measures may maintain the ecosystem stability. The increased leaching of N in the Dutch site is a major environmental concern. Removal of surface soil was proposed as a management measure directed towards counteracting increased effect of N-deposition. Another concern was the long term effects of drought on high altitude moors in e.g. Wales. The substrate will loose C and ability to retain water. This may result in increased floods after heavy rain or lack of water in drought periods. A possible action is to block drainage from the moors. A direct effect of changing climatic conditions were found on plants and soil in the beginning, but as time went on more indirect effects - effects that were more difficult to relate to specific causes - appeared. Finding good indicators is therefore complicated. Indicators can be related to processes or to eco-system conditions and can be indicators of change or indicators that explain changes. It is important that indicators are found that are unequivocal and relatively simple to measure, facilitating the monitoring of terrestrial ecosystems that are being developed in Europe in connection with the implementation of EU Water framework directive, Natura 2000 network etc. Several of the end-users are involved in the work with developing the monitoring program, thus the results will be exploited as input to indicator work and to make recommendations for monitoring. A website was established early in the project. The website is an important platform to reach larger groups of end-users. We have produced 5 web-based newsletters - also distributed to target persons and organisations in Europe and climate change fora. The content has been a mixture of scientific news, risk assessment and management. The last newsletter is a popular science leaflet as suggested by the end-users with a 2 pages summary of key findings from each of the six experimental sites and from the cross-site analysis of fauna and remote sensing. The scientific summary from each site can further function as a 2-pages handout to visitors at the sites. Scientific peer review publications are in preparation: a cross-site synthesis of shrubland ecosystems for a special issue as output from the AVEC workshop in Capri in October (submitted in February). Further, a series of papers are in prep. to go into a special issue of a scientific journal summer 2005. The further recommendations of continue presentations at workshops, conferences, meetings etc. will generally be in everybodies minds, including teaching in summer camps. The website will be paid and maintained by Claus Beier for another 5 years. This will serve as a mean to inform across the network, to keep the protocols updated across the groups and to share data and publications. This also includes common access of the partners to the data and linking to other networks.