Models for Adaptive Forest Management
FORSTLICHE VERSUCHS- UND FORSCHUNGSANSTALT BADEN-WUERTTEMBERG
Higher or Secondary Education Establishments
€ 700 600
Alfons Bieling (Mr.)
Sort by EU Contribution
STICHTING WAGENINGEN RESEARCH
€ 549 996,75
EIDGENOESSISCHE FORSCHUNGSANSTALT WSL
€ 469 800
EIDGENOESSISCHE TECHNISCHE HOCHSCHULE ZUERICH
€ 494 520
€ 507 684
UNIVERSITAET FUER BODENKULTUR WIEN
€ 648 000
EUROPEAN FOREST INSTITUTE
€ 763 515
€ 186 600
INSTITUTO SUPERIOR DE AGRONOMIA
€ 431 200
INSTITUT NATIONAL DE LA RECHERCHE AGRONOMIQUE
€ 273 600
€ 318 000
UNIVERSITATEA STEFAN CEL MARE DIN SUCEAVA
€ 125 760
€ 429 760
FORESTRY COMMISSION RESEARCH AGENCY
€ 403 787,50
POTSDAM INSTITUT FUER KLIMAFOLGENFORSCHUNG
€ 249 049,35
CENTRO DE INVESTIGACION ECOLOGICA YAPLICACIONES FORESTALES CONSORCIO
€ 96 000
IFER - Ustav pro vyzkum lesnich ekosystemu, s.r.o.
€ 138 000
University of Forestry
€ 64 800
Foreco Technologies S.L
€ 61 200
€ 49 200
Grant agreement ID: 226544
1 May 2009
30 April 2013
€ 9 070 933,80
€ 6 961 072,60
FORSTLICHE VERSUCHS- UND FORSCHUNGSANSTALT BADEN-WUERTTEMBERG
Forest management adapts to climate change
Grant agreement ID: 226544
1 May 2009
30 April 2013
€ 9 070 933,80
€ 6 961 072,60
FORSTLICHE VERSUCHS- UND FORSCHUNGSANSTALT BADEN-WUERTTEMBERG
Final Report Summary - MOTIVE (Models for Adaptive Forest Management)
The objective of MOTIVE is to provide insights, data and tools to improve policymaking and adaptive forest resource management in the face of rapidly changing climatic conditions.
Main S&T results/foregrounds
MOTIVE developed and evaluated strategies that can adapt forest management practices to balance multiple objectives under changing environmental conditions. The evaluation of different adaptive management systems took place within a scenario analysis and a regional landscape framework. The consortium developed a common understanding of the behaviour of standard decision makers reaching from a “no change manager” to the “forward looking adaptive manager” and formalized the decision-making process using a Bayesian update – approach.
An array of models (empirical as well as hybrid and process-based models) was employed in the analysis. Models were improved, for example regarding the simulation of disturbances under a changing climate, and further developed to model adaptive forest management regimes. The work on adaption strategies in the case study regions brought many improvements and new developments. For example, a new cutting-edge optimization algorithm was integrated into a complex mechanistic model for Mediterranean forests and helped adding economic aspects to the simulations in the Catalonian case study.
Besides the regional studies the MOTIVE consortium also worked on the European level. It developed insights into major trends for important climate parameters, growth, productivity and political developments. Based on a European species distribution model, a group of MOTIVE researchers was able to show that Climate Change may have severe economic impacts.
Answers to a questionnaire that was distributed to forests owners across Europe from Sweden to Portugal showed that the awareness of climate change is a crucial factor for adapting to climate change and that there is a clear decrease in the belief of forest owners in averse effects of Climate Change to forests from Portugal over Germany to Sweden.
The potential impact
The scientific work in MOTIVE was firmly based on participatory involvement of local and regional stakeholders and decision makers. These were instrumental in steering the model improvements and simulation studies in the case studies. A group of practitioners that formed the stakeholder advisory board accompanied the work of MOTIVE and formed a network that regularly met the scientists. This network will continue to exist beyond the duration of the project itself. The project also invested a lot of efforts into outreach to decision makers and politicians. The work of MOTIVE was published in two policy briefs and EFI organized a Think Forest event where MOTIVE results were presented to EU policy makers and stakeholders in Brussels. MOTIVE was also co-organizing a science-practice interaction session at the first European Climate Change Adaptation Conference in Hamburg. Furthermore, a decision support toolbox for adaptive forest management under Climate Change that contains crucial information and scientific results of the project has been developed. The website that hosts the toolbox will be maintained by one of the project partners and the framework of the toolbox will be used for follow-up projects.
Project Context and Objectives:
The ultimate objective of the MOTIVE project is to provide insights, data and tools to improve policymaking and adaptive forest resource management in the face of rapidly changing climatic and land-use conditions. Specifically, the following overall objectives are of primary importance for the integrated project MOTIVE:
• To provide an integrated assessment of forest management strategies that simultaneously considers multiple ecosystem goods and services rather than focusing on individual aspects such as timber production or biodiversity alone
• To provide a comparative assessment by using the same protocol across a wide range of climatic and socio-economic conditions by adopting a regional case study approach
• To consider the influence of feedback loops from socio-economy and policy making on resource management, thus leading to tools for adaptive resource management rather than to assume a priori defined management scenarios that are fed through a chain of models
• To translate scientific state of knowledge about expected climate change impacts into decision support for policy makers and forest practitioners taking into account uncertainties with respect to exogenous factors (climate, markets), so as to improve the robustness of the assessment tools in real-world decision making.
The general approach is that a limited number of selected simulation models at the stand, landscape and EU level are further developed around specific real landscape cases (see Figure 1). There are two cases per each of five bioclimatic zones and one more generic for the EU. This number of real cases is selected to reflect the diversity in European forests as well as the different functions they have and the main risk concerning them. All cases have some communalities such as the timber production and assessment of the impact of climate change on growth and competition, but each case also has specific characteristics, i.e. we will focus on certain case specific risk and management concerns, such as forest fires, coppice management, protection from natural hazards, optimized biomass production, or nature conservation.
1.1 Climate Change Scenarios
Niklaus E. Zimmermann, Dirk R. Schmatz and Achilleas Psomas
The global climate is currently warming and this trend is expected to continue towards an even warmer world, associated partly with drastic shifts in precipitation regimes (IPCC 2007). The global temperature has been warming by ca. 0.6°C (±0.2°C) during the 20th century (IPCC 2001), but the land areas have shown a higher increase in temperature within the same period. Here, we report on the current state of the art in climate model projections for Europe, with an outlook to the soon available 5th IPCC assessment report.
It is challenging to project how the climate might look like in 50–100 years, a duration that is relevant for forest management. In climatology many models are used in ensemble mode to generate possible climate futures. Each model and each simulation can be considered one possible representation of how the climate might evolve during the 21st century. For forest management and decision-making, we have to live with the fact that no exact forecast is possible. Rather, we have to implement our planning based on projected trends including their uncertainty. The periodic reports by the Intergovernmental Panel on Climate Change (IPCC) summarize the state of the art of how scientists see the development of the future climate and the associated impacts on ecosystems, economy and society. Now, the 5th assessment report is approaching, and some comparisons to the last two reports are already possible. The 3rd Assessment Report (IPCC 2001) had assumed that the global climate might be warming by 1.4–5.9°C, with no probabilities given for different increases, and with extreme scenarios projecting even far higher temperature increases. The 4th assessment report (IPCC 2007) provided a more narrow range of the likely future climate stating that temperatures will likely be between 2.0 and 4.5°C warmer than during the 1961–1990 period (with a likelihood of 66%). It also said that temperature increases by more than 4.5°C cannot be excluded (see Rogelj et al. 2012), but that the most likely temperature increase by 2100 is 3.0°C. First comparisons from global climate modeling studies for the 5th IPCC assessment report project an increase of 2.4–4.9°C as medians from three different scenarios of radiative forcing (following different emission scenarios that are similar to those used in earlier reports). A fourth scenario is added that assumes a more rigorous and rapid reduction of greenhouse gases than was ever used before, predicting a median temperature increase of only 1.1°C during the 21st century. Overall, the model simulations for the 5th IPCC assessment report expect that the likelihood of having global temperature increase exceeding 4.9°C is 14%, thus also likely, but that the most likely warming scenario at the global scale is still 3.0°C. Thus, in general, the newest scenarios do project similar average warming trends as we have seen in the 4th IPCC assessment report, although some scenarios point to somewhat higher warming trends than were calculated for the 4th report (see Rogelj et al. 2012). Figure 1 shows global climate data simulations for the 4th and 5th assessment report.
The global climate is simulated using so-called general circulation models (GCM), which project the climate future on physics-based processes and first-principles. For regional applications such as e.g. Europe, such GCM model output has a too coarse spatial resolution, usually in the range of 1°–2.5° Lat/Lon per model cell. In order to obtain more realistic climate projections at a regional to local scale, two types of downscaling are often combined. First, so-called regional climate models (RCM) are calculated to certain larger regions of the World (e.g. all or parts of Europe). These models contain the same physical mechanisms as the GCMs, are fed by GCM output, and simulate the climate development within the study region by using GCMs data as boundary input to the study region. The output of these models is at high temporal and moderate spatial resolution, ranging typically between 5-50 km per cell. This is a much better spatial representation of the climate in regions and the output is somewhat sensitive to mountains and their effects on the climate system, though often the output is still too coarse for management and decision-making. Therefore, a further statistics-based downscaling procedure is applied (Pielke and Wilby 2012) in order to scale the output from RCMs to finer spatial resolution ranging from e.g. 100 m to 1 km, which can be considered well-suited for management applications.
Climate projections for the MOTIVE project
For the MOTIVE project, we have used five different RCMs driven by four different GCMs resulting in six GCM/RCM combinations in order to study the impact of likely climate changes on forest species and ecosystems. Table 1 gives an overview of the models used, which originate mostly from the ENSEMBLES EU project, using GCM runs that were calculated for the 4th IPCC assessment report (IPCC 2007).
We down-scaled basic RCM output variables such as monthly temperature and precipitation to finer spatial resolution, typically to 1 km or 100 m cell size. The method used can be called the “anomaly-approach”, where we scale the deviation of the future compared to current climate from coarser to finer resolution. This is an efficient method, since anomalies do not depend much on altitudinal lapse rates. Once downscaled, the anomalies are added to an existing high-resolution climate map such as those available from Worldclim (Hijmans et al. 2005) or from national mapping campaigns (e.g. Zimmermann and Kienast 1999). The most important step here is to generate anomalies appropriately. First, we need to know the reference period of the high-resolution climate maps. Worldclim is mapping e.g. average monthly values of the 1950–2000 period. Next, we generate the monthly climate anomalies for given periods in the future. To calculate the anomaly of each projected future climate month of any RCM relative to the current climate, we use the simulated time series outputs for the re-analysis period of 1950–2000 from each RCM. By this, we avoid the projection of modeled bias in RCMs should the recent past be wrong compared to climate station measurements. We are only interested in projecting the relative difference between simulated recent past and simulated futures. Once anomalies are generated, we interpolate these anomalies to the high resolution of existing climate maps such as Worldclim and add them to these maps to project the future climate changes to the representations of the existing climate.
The development of climate anomalies was done by (a) first averaging the monthly time series of minimum (Tmin), average (Tave), maximum (Tmax) temperature and precipitation (Prcp) over the period of 1951–2000 for each RCM run used, since these represent the same base period of Worldclim maps. Then we used monthly RCM outputs to calculate monthly anomalies relative to the 1950–2000 period means per month. We developed monthly anomalies: (a) by subtracting current from future climate for temperatures, and (b) by dividing future by current climate for precipitation. The latter results in ratios of change, which avoids negative precipitation values that could else result after down-scaling if the difference method is used. All climate anomalies are first calculated at the spatial resolution of the RCM output, and is then scaled the medium resolution of 1 km by a bilinear interpolation (and in a second interpolation step to 100 m if necessary). Figure 2 illustrates the projected climate change trend from the six used RCM simulations by the example of the annual and seasonal (summer and winter half) means, and by the uncertainty in projected summer climates.
For temperature, we observe a general warming trend in the range of 1.1 to 4.1 °C, with least warming in the winter half in Atlantic regions (+1.5°C), and highest warming in the Boreal North (winter; +4°C) and Mediterranean South (summer; +3.5°C). The Alps generally face higher warming trends than the surrounding mainland, specifically in the winter months. The uncertainty among the six models is lowest in Southwestern Europe, increases towards North and East, and is highest in Eastern Europe. For precipitation, the trends show a similar and even clearer segregation between North and South. Winters are projected to become significantly wetter in Northern Europe (+30%), and to a lesser degree also in Central Europe (+15%), while Southern Europe is projected to become slightly drier (-15%). Summers are projected to become significantly drier in Southern Europe (-35%), and to a lesser degree also in Central Europe (-20%), while Northern Europe is projected to become slightly wetter (+20%). The uncertainty among the six models is highest in the (sub-) Mediterranean regions, in the Alps and in the far North of Europe.
General implication for forest management
For forest management, the projected climate anomalies may require specific actions to avoid significant loss in timber value. Least changes are likely required for (the far) Northern Europe. Here, the evaporative demand of a warming climate is balanced by higher precipitations both in winter and summer. Forest productivity can be expected to increase, and more thermophilic species may soon find suitable habitats in this region.
For Central Europe, the projections are still quite unclear. While there is a general warming trend projected, the models disagree as to the magnitude of warming, and whether precipitation will increase or decrease. But even if no changes in total precipitation amount will occur, there are likely to be two effects relevant to forest management. First, evaporative demands due to warmer temperatures can likely not be fully balanced, specifically because summers become slightly drier, the result will be a net water loss for tree growth; the general tendency is a climate seasonality shift towards a more Mediterranean-type climate, away from a summer maximum and winter minimum in rainfall towards two rainfall peaks in spring and fall, with comparably dry summers. In some regions (especially towards the Atlantic coast), this trend is less pronounced, and both the changes in rainfall and in temperature are dampened by the proximity to the ocean.
Most severe changes with negative consequences for timber production can be expected for the Mediterranean region and its neighboring areas in Southern Europe. Here, precipitation is decreasing both in summer and winter, and temperatures are increasing in both seasons more (winter) or less (summer). This will result in much drier growth conditions, and is likely to have severe effects on the already water-limited forests. Only (and not shown in figures), the temporal variability for this region is specifically high with regards to rainfall, so that we may still expect some wet years in-between very dry years. This may mean that natural forest regeneration may still be possible.
While the general climate trends are still uncertain, as seen from the uncertainty maps originating from 6 RCM models, projections of climate extremes are even more difficult to make or to foresee. Several models, and even more so the deviation among models, are projecting that both the climate variability and the uncertainty of projections will become larger towards the end of the 21st Century. In general, we can expect that both temperature and precipitation extremes will increase. This has the consequence that forest management becomes more difficult, because a larger range of possible conditions will need to be considered in the planning and decision-making.
1.2 Future Ranges in European Tree Species
Niklaus E. Zimmermann, Signe Normand, Peter B. Pearman and Achilleas Psomas
Climate is a major driver of plant and tree distribution, while soil variables or inter-species competition are of¬ten considered to primarily drive their local abundance. The climate constraints to species ranges are generally accepted. Therefore, a changing climate is especially relevant to long-lived plants such as trees or shrubs, as these take many years to reach maturity and, given their long life and stationary nature, they are especially vulnerable to rapid changes in climatic conditions. In addition, forest management typically encompasses many decades, partly even reaching to the end of the 21st century, which illustrates the challenge to manage such or¬ganisms successfully at such long planning periods. This calls for careful and adaptive management strate¬gies and for a good understanding of the uncertainties related to the expected changes and their impacts on trees and forest ecosystems.
Many approaches exist to project the impact of climate change on trees and forests. Most of these ap¬proaches can either only be applied to comparably small regions or too few species, or they need to be run at very coarse spatial resolutions in order to cover larger areas such as Europe. The following five basic ap¬proaches can be distinguished: (1) biogeochemistry models, (2) population dynamic models with competition (3) demographic models of single species; (4) phenological models of single species; and (5) species distribution models. The model types 1–4 are usually more process-oriented than type 5, and therefore contain biological realism in what they project under climate change. Their general limitation is usually in that they are not capable of simulating the future fate of species at large spatial extent (such as Europe) and simultaneously at a comparably fine spatial resolution (such as ≤1km grids) that is useful for forest manage¬ment. Several of these approaches (3,4) lack the capacity to include competition among species, while almost all approaches (except few models of type 1–3) actually include seed dispersal and thus can provide insight into natural migration rates following climate and land use change.
Species distribution models (SDM) of the last approach (5) are most often used to project climate change effects on the suitability of an area for a given set of species. This represents a comparative method that relates the observed distribu¬tion of a species to the environment (such as climate, topography, soils), and calibrates statistically the ranges of species as a function of these environmental predictors. The method is capable of simulating large spatial extents (such as Europe) at a fine spatial resolution (≤1km), and once calibrated, it can be rapidly applied to changing environmental conditions. It thus provides useful information for forest and conservation managers in their decisions to cope with global change. However, it is important to also consider the limitations that go with this approach. First, the method is not dynamic, and only provides information of habitat suitability under future environmental conditions. It therefore cannot foresee by when a species naturally will invade an area. Although, several novel approaches have now been developed to combine SDMs with realistic migration simulations (e.g. Meier et al. 2012). Second, the method is not based on physiological first-principles, and thus is an empirical, not a mechanistic approach. It is therefore not fully reliable when projecting to very novel climatic conditions that cannot be observed today. This limitation is, however, valid for several of the above-mentioned five approaches.
Species distribution modelling
In the following, we present SDM simulations, often also referred to as climate envelope models (CEM), for major tree species in European in order to assess what the consequences of climate change on the habitat suitability of these tree species might be. We used presence/absence information from the ICP Forest Level I inventory (Lorenz, 1995) in order to build a database of tree species presences and ab¬sences across Europe. We compiled data for 38 tree species at a total of >6,000 inventory plots. We then compiled a series of climate maps under current and potential future climate from downscaled RCM models for future climates (see Chapter 2). Additionally, we compiled some topographic variables that additionally may influence the spatial patterns of trees. Prior to selecting the variables, we executed a varia¬ble importance analysis for each tree species separately. This was done in order to adjust the variable selec¬tion to those that have a strong effect on the range dynamics of the species. We refrained from using all possible climate variables on order to avoid too high correlations per species, and in order to keep control over the number of variables we maximally allowed entering the models.
We finally designed the variable pre-selection as follows. First, no more than five variables were allowed per species. Second, for species with less than 50 observations, a minimum of ten observations per variable added was required. Third, among similar and highly correlated variables, we selected variables that (a) showed high predictive performance, and (b) did correlate >0.7 to any other variable selected. We only modelled species with at least 25 presence observations, meaning that such a species would only be fitted against two predictor variables. Most species were fitted against five (24 species) or four (12 species) variables. Four and three species were fitted against three and two variables, respectively, due to the low availability of presence points in the ICP Forest Level I data set.
We selected predictor variables from the following groups of environ¬mental predictors: (1) temperature – either degree days with a 5.56°C threshold or minimum temperature of the coldest month; (2) precipitation – we computed seasonal means and selected among those per species; (3) moisture index (difference between potential evapotranspiration and precipitation) – we computed this index for spring (March–May) and summer (June-August) and se¬lected one of the two according to its predictive power; (4) poten¬tial global radiation – we computed winter or annual mean radia¬tion values and selected according to predictive power; (5) slope angle in degree – we added this variable, if it was among the top 5 uncorrelated variables for a given species. By this we allowed to simulate habitat suitability based on predictors that are (a) relevant for a given species, (b) not highly correlated, and (c) we did not include too many variables into models, which might cause prob¬lems in a non-analogue climate due to changes in correlation among current to future climate variables. Winter and spring condi¬tions were generally speaking more important for Mediterra¬nean trees, while summer conditions were more powerful predic¬tors for Central and Northern European trees.
Potential future climate was taken from six different RCMs (see chapter 1), providing a range of potential climate futures. The use of several RCM models provides the mean trend that can be expected from climate change impacts on trees, and it allows us to derive a measure of uncertainty associated with the projection of these trends (Thuiller et al. 2009). Several statistical models were used, since the choice of a specific model has been shown to signifi¬cantly contribute to uncertainty in projections. Therefore, given the use of six statistical models and six future climate model runs, we model 36 different possible futures per species and time slice. This allows for assessing the projection uncertainty from both the variability in climate models and the variability originating from the choice of statistical methods (see Figure 1).
We optimized each statistical model following procedures de¬scribed in Thuiller et al. (2009) and where feasible, we maximized kappa to select a threshold to split probabilistic projections of species presences into simulated presence and absence values. We therefore had one presence/absence map per climate model/statistical model combination available. We then built ensem¬bles of these model projections and classified these as follows: (1) a species is unlikely to find a suitable habitat if less than 30% of the projections indicated presence of a species; (2) a species is moderately likely, associated with high uncertainty, if 30–60% of the projections suggested that the species is there; (3) a species is most likely present with rather low uncertainty under projected climates if in >60% of the 36 model projections presence of a species is simulated to occur. Such a simple classification avoids an over-interpretation of the re¬sults from the simple model approach.
Figures 1 and 2 illustrate the potential future range shift in two spe¬cies, namely Picea abies (L.) Karst. (Norway spruce), which is a species of Central to Northern origin, and Quercus ilex L. (holm oak), which is a species of Mediterranean origin. The first species (as a common Central to Northern European species) is expected to lose much terrain at low altitudes and in Central Europe, and will retract to higher alti¬tudes and latitudes following climate change. Currently, Norway spruce is planted at lower altitudes than it would naturally occur. Obviously, these lower altitudes are still within the fundamental niche of the species now, and the distribution model under current climate does include these areas as suitable habitats. However, much of this area will soon become unsuitable. This is specifically visible for the simulations of the period from 2051–2080, where for Norway spruce clearly only higher elevations and latitudes are projected to remain suitable. Larger parts at low altitudes and Central Europe become unsuitable or remain only suitable with high uncertainty. This uncertainty arises from highly contradicting projections by both climate and SDM model combinations.
For holm oak, the picture is very different. This evergreen species is adapted to dry and warm climates and with¬stands the repeated climate extremes by growing slowly and investing non-structural carbon into leaves, bark and roots to avoid drought damages, but does so at the cost of a slow maximum growth. The species is projected to extend its range considerably towards the North. E. g. these models project suitable habitats along the Southern Atlantic coast of France, where the species has been observed to natu¬rally extend its range from initial populations recently.
All 38 tree species have been simulated for Europe, and only two species are displayed here. A more complete set of species can be checked out and downloaded from a dedicated website . It becomes obvious that mostly the more drought-tolerant species such as sessile oak (Quercus petraea), pubescent oak (Quercus pubescens), or Scots pine (Pinus sylvestris) can be expected to become more abundant at lower altitudes through¬out Europe, while other species such as common beech (Fagus sylvatica), sycamore maple (Acer pseudoplatanus), lime (Tilia spec.), elm (Ulmus spec.) or silver fir (Abies alba) are likely further reduced in their ranges similar to Norway spruce. Species from (sub-) Mediterranean regions such as holm oak (Quercus ilex), hop hornbeam (Ostrya carpini¬folia) or cork oak (Quercus suber) are expected to extend their ranges to the North, but these species will not reach the areas currently suitable for beech or spruce by the end of the 21st century. Different pine species are also expected to extend their ranges quite considerably. However, they will likely not extend to very fertile soils either, and some of the species like e.g. Scots pine (P. sylvestris) might face indirect threats from insects and other pest outbreaks, rather than direct threats from climate change alone. In summary, the projected range shifts will affect the forest structure quite considera¬bly. Among all species modelled, we can expect a shift in plant functional types in Europe be¬tween now and the end of the 21st century (see Figure 3).
Discussion and conclusions
Such changes will affect the functioning of forest ecosystems and the services we can expect these ecosys¬tems to provide. One study (Hanewinkel et al. 2013) found a likely severe loss in timber value resulting from such tree species habitat suitability shifts for most of Europe, except for most of Northern Europe. This is due to a loss in suitability for valuable timber trees such as Norway spruce or beech at the cost of less valuable trees with slower growth potential, such as sessile or pedunculate oak.
How reliable are such model projections? The displayed maps only represent the habitat suitability by a certain time period (e.g. 2051–2080), but do not predict that a species will disappear or invade as rapidly as displayed by the suitability maps. The response of species will in most cases be much slower (see Meier et al. 2009 for examples and related discussions). If climate is simply shifting its means, and no climatic ex¬tremes occur, then most likely the natural re-adjustment of species ranges will take considerable time, ranging from Centuries to Millennia. However, there are two reasons that may explain faster responses. First, although only minor range shifts have been observed in trees so far, we can expect an acceleration in the range shift response to on-going climate change in the near future. Until now, the degree of warming has not clearly exceeded the natural range of variability a tree species experiences at any given location. This range will be exceeded in most locations by 2050 due to an increasing warming trend. On the other hand, most scenarios project more frequent climatic extremes, and an increase in climate variability, resulting in even more severe extremes. Such extremes have been shown to affect species ranges (Zimmermann et al. 2009), and they will cause severe effects specifically at the rear edge, where climate is hot and becoming drier in Central and Southern European summers. This means it is likely that the northward movement will naturally occur at a steady pace with smaller forward leaps, while the contraction at the rear edge of species ranges may occur in more acute events following climatic extremes and the subsequent pest outbreaks.
1.3 Adaptive potential: A partial insurance against climate changes risks
Jean-Baptiste Lamy, Sylvain Delzon and Antoine Kremer
To exploit natural adaptation to climate: a weapon to cope with climate change
Adaptation of forests to climate change can occur via inherent processes or may be triggered by measures instigated by humans. The rationale for considering inherent physiological processes comes from the evidence that tree populations have undergone profound genetic changes during the natural warming after the last glaciations facilitating their adaptation to changing climate (Petit et al., 2008). Provenance refers to the specific geographical location that marks the natural origin of a tree. Natural selection by evolution has adapted each provenance to its local environment, hence, there are genetic differences between different provenances of the same tree species. In provenance testing, seed is collected from several provenances and planted for comparison in random replicated experiments at various forest locations (Figure 1).
Since provenance tests were first installed by forest practitioners, results accumulated showing substantial population divergence for almost any adaptive trait that has been investigated so far. During the past decade, the evolutionary historical trajectories during the Holocene (our current interglacial climate period) have been reconstructed for most of the European tree species, providing some clues about the rates of evolutionary changes. A prevailing view resulting from combined genetic and historical investigations is that adaptation can occur at rapid time scales – even contemporary time scales- provided that there is enough genetic variation. The issue of future adaptation to on-going climate change can therefore also be thought of in evolutionary terms.
The fate of extant (or living) tree populations undergoing severe environmental changes is related to their adaptive potential. Practitioners are seeking studies which imitate climate change so they may evaluate adaptive potential that would guide their management options. Evolutionary scientists are attempting to identify ecological and genetic drivers or processes contributing to the adaptive potential.
Adaptive potential of a tree population can be defined as its capacity to respond to a given environmental change, by modifying its own genetic composition and/or by modifying its phenotypic expression. In more scientific terms, adaptive potential is the sum of the changes due to genetic adaptation and changes due to phenotypic plasticity .
Measuring adaptive potential: response and transfer functions
Adaptive potential in tree species is assessed by monitoring the same provenance in different ecological settings (see Figure 1), for instance by planting a “northern” provenance more in a “southern” provenance test location. This approach is relevant in the context of climate change especially if environmental changes over spatial gradient can mimic future climate changes.
Adaptive potential is, then, described by a response and/or transfer function. The response function describes the provenance’s performance along a climatic gradient, indeed data is needed from several provenances tests (>6) with the same populations in each. These experiments are laborious, costly, and only available for some valuable species as pine, Norway spruce or beech. For the majority of tree species, there are few provenance experiments with many provenances tested.
A valuable alternative strategy is to use a transfer function i.e. the provenance’s performance (or standardized proxy) is described along a transfer distance (see below). This method relies more on assessing contrasts between populations at the same site, rather than contrasts between same populations at different sites. Here, we present a statistic free approach to illustrate this concept.
For each population the transfer distance (ΔE, see Figure 2) was calculated as the difference between the mean of a given climatic variable (usually temperature) at the testing site (provenance test) and the mean of the same climatic variable at origin of the provenance. When temperature is used as the climatic variable, negative values on the x axis can therefore represent transfer to warmer climates (ΔE < 0), while positive values indicate transfer to cooler climates (ΔE > 0). Each population performance was also standardized by the local population performance (ΔG, see Figure 2). A negative value of ΔG means that populations performed less well compared to the local population and positive values mean that the foreign population outperformed the local population. A species response function with a positive slope, as it is illustrated in Figure 3.2 means a transferred provenance in a warmer climate could increase provenance performance when compared with local provenance.
Re-analysis of old Quercus petraea provenance tests
Almost all available provenances tests were not designed to mimic climate change. However, the comparison between the transfer distance realized in four French provenance tests (with more than 100 provenances tested) on Quercus petraea and the expected climatic change at the provenance origin showed that provenance tests are valuable tools for prediction about provenance performance in the climate change context.
Figure 3 shows how the transfer distances realized in the French provenance tests overlap the expected climatic change at the provenance origins, however, only the transfer to warmer climates are useful (transfer distance > 0). For example, a Western German provenance planted in a Northern French test location simulates a rise in temperature of 1.5 °C, Such an increase in temperature is expected in Germany by 2040. Using the described strategy for old data from Quercus petraea, we adjusted a species transfer function (see Figure 4).
As seen in Figure 4, most populations show positive transfer distances, meaning that they were moved on average towards warmer climates. The y axis is the difference in height between the transferred population and the local population (at the site where the test is established). All populations above the 0 on the y axis are taller than local populations. The transfer function has a negative slope, suggesting that on average transfer to warmer climates is likely to reduce the tree height. An increase of 1°C in the mean annual temperature decreased tree height by 12.3 cm at 10 years old. However there was a very large population variation around the mean response. At least 8 populations (see populations above the confidence interval) actually will grow better than the local population when they are transferred to warmer climates.
Such approaches, in order to accelerate forest adaptation, are under evaluation in North-America, and future seed transfer guidelines will take in account such considerations. In contrast European countries are just starting such work despite large numbers of provenance trials and an old tradition in forest science. These genetic approaches coupled with adaptive management strategies could buffer or maintain sustainable levels of forest productivity and ecosystem health.
1.4 Mapping the Risk to European Forests with a Changing Climate
Barry Gardiner, Mart-Jan Schelhaas and Bruce Nicoll
Insects and diseases, storm damage, wildfires, drought, and herbivore grazing are the main agents affecting forests within Europe although their importance varies by region. The level of damage is increasing at the same time that European forests are being asked to provide an increasing range of ecosystem services and therefore understanding the impact of forest management and the changing climate on forest disturbance is one of the big current challenges for forest science.
European Forest Disturbance
The primary abiotic damage in Europe is due to wind and fire and the primary biotic damage is due to bark beetles. However, these damage agents are quite localised so wind tends to be more important in Western and Northern Europe and fire more important in the Mediterranean region, and beetle damage is important in Atlantic temperate, Alpine and Mediterranean forests, whereas mountain and boreal forests are more affected by fungi and wildlife. The economic consequences of this damage to forests can be profound: for example fire caused €5 billion of damage in Greece in 2007 and €0.8 billion in Portugal in 2005, and wind caused €6 and €1 billion in France in 1999 and 2009, respectively, and €2.4 billion in Sweden in 2005.
The overall increase in European damage levels appears to be due primarily to an increase in the forest growing stock together with changes in climate and land management (Figure 1). Such changes are expected to continue in future. For example, wind damage is predicted to increase by at least a factor of 2 by the end of the century and extend further east across the continent. In addition, warmer and drier summers will increase fire risk. For example, in France the high levels of forest fire danger currently limited to the Mediterranean area, will extend to the western part of France by 2040 and to most of the country by 2060. Global warming is expected to enhance the winter survival of bark beetles triggering increases in population abundance and risk of outbreaks.
European forests will, therefore, be under increasing threat from a range of existing and new damaging agents and it is important to be able to predict the level of risk now and into the future. Furthermore, because major damage events and outbreaks of pests and diseases can affect large areas and many countries such calculations need to be carried out from the regional to the European scale.
Modelling Disturbance at the Landscape to European Scale
It is now possible to work at regional to European scales without losing sight of the detailed mechanisms controlling tree damage which operate at the tree scale or lower. This is because of recent developments in computer science, complex system modelling and artificial intelligence. However, high-resolution modelling of risk to forests from different hazards at the European scale presents serious challenges. In particular there are requirements for detailed data on forest structure (species mixture, tree height, diameter, spacing, ground vegetation, etc.) and site (soil type, elevation, climate, etc.) and to have available models that are able to calculate risk across Europe for the range of site types, forest species, forest management regimes and climate that occur. For illustration, we present here modelling of the risk of wind damage and bark beetle attacks across Europe.
Mapping of wind damage risk for Europe
To map wind damage risk across Europe requires knowledge of the current and future wind climates, information on the forest stands (tree species, tree diameter and height, stocking), and soil type and rooting depth. In this study we used high resolution wind climate data for the past and future climate that is available at 25km resolution from the EU ENSEMBLES project (Figure 2). Information on the forest structure was obtained from the Synthetic European Forest Structure Database, which has a resolution of 1 km and is based on a species distribution map of Europe and a collection of National Forest Inventory (NFI) measurements. The FAO soil map that was used to construct the species map was also used to assign soil type and rooting depth. The stand and soil data described above were used as inputs to the wind risk model ForestGALES2.3 which calculates the critical wind required to damage a stand. The output from the ForestGALES model run is shown in Figure 3. Only areas where the land cover is more than 10% forest are shown. From this figure it can be seen that there are areas of Central Europe, southern Sweden, Southern Finland and Estonia which have the lowest critical wind speeds. If the critical wind speeds are combined with the wind climate from Figure 2 it is possible to calculate a probability of damage for the whole of Europe and data on future wind climates allows the future risk to be calculated.
Mapping of bark beetle risk for Europe
Tools are not currently available that allow mapping of the risk of bark beetle outbreaks across Europe. However, scientists at Boku University in Vienna performed a large number of simulations with the PICUS forest growth model, covering a wide range of stand and climatic conditions in Austria. These simulations were then used to construct a regression model that predicts the chance of bark beetle infestation based on average annual temperature, total annual precipitation, stand age, stand density and the percentage of the host tree within a stand. If a stand is infested the damage level is then predicted as a function of average annual temperature, total annual precipitation and host tree percentage.
We applied this regression model to the whole of Europe, using the Synthetic European Forest Structure Database (see above) and WorldClim data focussing on spruce. Stand age was estimated from height and soil type, based on yield tables. Climate data were derived from the WorldClim database, as average annual temperature and average annual precipitation. Stand risk was quantified for each cohort in the forest structure database as the product of total volume of spruce, the chance of infestation and the fraction of the stand damaged. The resulting map (Figure 4) gives the endemic risk of bark beetle infestation, so it ignores increased risk due to build-up of the population in previous years, for example due to wind-throw events. Furthermore, this reflects the average long-term weather conditions. As can be seen from the map Central Europe and Southern Sweden are major areas at risk of bark beetles on spruce, which corresponds well with observed damage outbreaks. Although spruce is also found in more northerly areas it is too cold at these latitudes for bark beetles although this could change in the future with a changing climate.
Summary and Conclusions
We have shown that it is possible in a preliminary way to calculate the risk for wind damage and bark beetle attack across the whole of Europe. Such calculations require extensive data sets on European forest condition, and information on the current and future climate. In addition these maps can only be created by models that are able to work across the whole range of forest and site conditions that exist within Europe. The use of informed decisions on how to adjust these models when data are not available for a particular species or site is also an important requirement. Therefore, these maps have to be treated with caution and we emphasise that at this stage they only demonstrate the possibilities for the future and give some indication of the most vulnerable areas of Europe to these hazards.
With improved data on European forests and soils such as is becoming available from remote sensing it will become possible in the future to put higher confidence in the outputs from such simulations. At the same time it is important that work continues to extend the existing risk models for different damage agents and to start building models for other important hazards such as forest fires for which at present only locally applicable models are available.
1.5 Adaptive forest management
Harald Bugmann and Antoni Trasobares
Forest managers have a long tradition of adapting their silvicultural practices to new insights gained upon past experience. Anthropogenic climate change, however, is proceeding at a magnitude and speed that is unprecedented in the history of human civilization, and forests will likely be out of phase with climate over the coming decades to centuries. Thus, information on the likely future development of the drivers of forest dynamics, such as climate, and the resulting impacts on ecosystem properties and ecosystem services must be taken into account in forest decision making already now, rather than in hindsight only. This means that a pro-active, forward-looking approach is needed for managing forests.
The forest models developed in the MOTIVE project are thus essential for projecting future forest dynamics to support current-day decision making. The ‘decision space’ of potential forest management actions was determined by using the MOTIVE forest models for simulating a wide range of management practices, including current practices as well as adaptive, forward-looking management regimes. For each case study region, this was done for a set of forest stands in the framework of a set of specific management objectives that aimed at maximizing the goods and services from the stands, while minimizing risks. Also, a range of climate scenarios was taken into account.
The goals of these analyses were as follows: (1) to assess the likely changes of forest stand dynamics and of the provision of relevant ecosystem goods and services for the major European forest ecosystems; (2) to assess the utility of adaptive vs. conventional management regimes for maintaining the provisioning of forest ecosystem goods and services; and (3) to allow for cross-case study comparisons of the results regarding the effects of adaptive management.
In a first step, pilot studies were conducted to assess the concept of adaptive management underlying MOTIVE (Figure 1), showing that it was appropriate and efficient. These first results were already consolidated for some case studies but incomplete for others.
In a second step, refined simulation studies were conducted, again on a per-case study basis, yielding significant and novel scientific results on (1) climate change impacts on forest ecosystem goods and services, and (2) the identification of the most suitable management strategies. Generally, the interactions between forest management (both current and adaptive strategies) and climate change were elucidated, and the effects on the provisioning of a given set of targeted ecosystem goods and services were analyzed through time. The following chapters will present exemplary results from all the case studies.
As a prerequisite for the robust application of the MOTIVE forest models under a wide range of scenarios of climate and management practices, they were significantly improved and further developed to allow for a better representation of the ecological processes underlying forest dynamics and an advanced evaluation of adaptive management strategies (e.g. improving climate sensitivity, more accurate calculation of ecosystem goods and services).
Examples of model improvements are the refinement of the species suitability model used in the Welsh case study, a more realistic module for ungulate browsing introduced for the Austrian case study, the coppice management module that proved to be essential for the Bulgarian case study, or a novel bark beetle module developed for the German case study.
Innovative methods and tools were implemented for the optimization of adaptive management at both the stand level (e.g. simulation-optimization approaches in the Finnish and Spanish case studies) and the landscape level (e.g. landscape level optimization developed for the Finnish case study). These methods and tools were developed in close interaction with key stakeholders in the case studies and thus they should be of practical relevance.
Below, we highlight three different challenges that researchers were faced with in the context of adaptive management in MOTIVE, and how they solved them.
First, it is not trivial to answer the question whether “conventional” management may suffice in the face of climate change, or whether a truly forward-looking, adaptive approach is superior. In the Welsh case study, several management approaches were set up for investigation including: 1) a business-as-usual approach; and 2) an adaptive species-diversification approach. Simulations showed that decision-maker 1 could be facing worse results in terms of the provision of ecosystem goods and services and the aversion of risks by postponing adaptation. The diversification management approach resulted in much closer trajectories (i.e. less uncertainty) between climate model variants than BAU. BAU management had possibility of higher economic returns but also of much lower returns.
This example highlights the importance of a key question in the current discussion on adaptive forest management, and shows that it can be answered by systems analysis: under which circumstances and at what point in time is it most beneficial to adapt?
Second, severe environmental limitations, e.g. by drought in Mediterranean countries, may call for specific adaptation measures. The results obtained for Mediterranean Catalonia, where a multi-objective stand-level optimization for P. sylvestris forests was developed and applied, and for Mediterranean Portugal, where cork oak management was simulated, clearly illustrate the effects of severe drought and suggest management strategies for adapting to this driving factor. These case studies are excellent examples of a ‘live laboratory’ of phenomena that will emerge in other parts of Europe in the coming decades.
Third, when implementing a given adaptive management regime (e.g. for even-aged management) at the landscape level, its effect and efficiency may be highly conditioned by the initial state of the forest, as expressed e.g. via the diameter distribution of the set of stands that compose the landscape (Figure 2). This is particularly pronounced in landscapes that are composed of many stands of similar ages, resulting from land-use legacies such as increased afforestation in the late 19th century or after the Second World War. Thus, the most appropriate management of forests does not only depend on the likely future trajectories of climate, the most limiting abiotic factors, and the specific ecosystem goods and services that are demanded by society, but also by the legacies of past uses of the landscape, thus sometimes strongly reducing the “manoeuvring space” of current and future management.
Overall, the results of the case studies demonstrate that the MOTIVE project has achieved robust simulation results of climate change impacts on stand dynamics and key ecosystem goods and services for the major European forest ecosystems, including approaches for adaptive management to alleviate negative consequences of climate change.
1.6 How fast can European forests adapt to a changing climate?
Geerten Hengeveld, Mart-Jan Schelhaas, Christopher Reyer, Niklaus E. Zimmermann, Dominique Culmann, Gert-Jan Nabuurs
The large diversity in abiotic and biotic circumstances in European forests makes it extremely difficult to predict what the impacts of climate change will be on the various tree species, and ecosystems at the various localities. This makes it even more difficult to analyse how forest management should adapt in order to take the changing circumstances into account at the right time and at the right pace. The case studies in Motive provide however a basis for upscaling to the European scale. For the first time we combine here species changes as predicted by a climate envelop model, with an incorporation of forest management responses in an empirical European forest resource model (EFISCEN).
This combination allows at the European scale to estimate how fast forest resources will change, under the assumption that existing trees on a site are plastic enough to survive the climate changes until the end of their normal rotation. It is assumed that only then a forest owner will decide to change tree species at that site towards one that is more preferred according to the climate envelop model. The owner will do this through shortening the rotation by approximately 10 years for susceptible species in order to speed up conversion towards more preferred species. This gives insight in fulfilment of raw material supply, forest resources, tree species, and increment under this adaptive management.
The results indicate that tree species composition will change only slowly at the European scale (see figure 1). By 2070, 10% of the total forest area will have changed species if species change at rotation end follows the climate envelope models. This can be increased to 12% by adaptive management, anticipating at expected species shifts. This is respectively 20% and 23% of the area change that is indicated by the envelope models. Large differences occur in Europe, with Northern Europe and Central Western Europe showing a higher rate of adaptation and especially South Western Europe a slower rate.
Overall, increment increases under climate change as compared to current climate due to positive production effects especially in Northern Europe (see figure 2). However, climate change effect is negative in South Western Europe. Adaptive management slightly reduces increment as compared to current management under climate change due to a higher share of temporarily slower growing forest (forest under regeneration). Raw material supply is not affected by climate change or adaptive management in this modelling study.
1.7 Climate change and the practical forest management: When to worry and when not?
Jette Bredahl Jacobsen, Rasoul Yousoufpour and Bo Jellesmark Thorsen
Across Europe climate change is expected to have quite different impacts on forest ecosystems varying with e.g. latitude, altitude, local ecological conditions and forest management practises.. However, just how large and how severe the changes will be is subject to considerable uncertainty. Therefore a crucial question is when to adjust to such long-term expected changes. Should adjustment be made now in anticipation of change but with great uncertainty about changes and impacts, or made only when changes and their impacts are observed and understood better, or should they be made gradually as we see the impact on forests? To what extent is it relevant and possible to go for strategies that try to optimize expected outcomes and make the most of forthcoming changes? When might it be equally relevant to pursue strategies that focus on reducing, as far as possible, the potential losses, and hence maximize the outcome in the worst case?
This chapter discusses these issues using two stylised approaches to handling the uncertainty forest managers face. These approaches we call the reactive approach and the proactive approach. We discuss when the benefits and importance of encouraging a proactive decision making approach among forest owners is particularly large, and when a more reactive management approach might be favourable or at least justified.
The reactive management approach
The reactive manager can be described as a decision maker who awaits and observes the actual outcomes and impacts of climate change as it develops, and adjusts management gradually to observed effects only. This decision making approach does not include forecasting or forming expectations about the likelihood of different climate change developments, nor their potential future impacts on forest ecosystems. A decision maker using this approach does not adjust current decisions to the possible implications indicated by such forecasts or expectations.
The results of forest owner surveys undertaken in MOTIVE indicate that while many forest owners do consider climate change a factor in their decision making, others are less explicit about this. Decision making approaches like this may be quite widespread among forest managers. If uncertainty about direction and/or impact of climate change is very large, the decision maker has little basis for firm expectations, and it may be a relevant approach if not the most favourable one. It may also be a relevant approach to follow if the expected changes are small, or if they are of gradual form. This could be where changes mainly affect growth or other ecosystem aspects in a way where management schemes like thinning or regeneration decisions can be adjusted smoothly and gradually fitted to the changed forest state.
Among the MOTIVE case studies, we do find specific cases, where such an approach is at least close to optimal. The Boreal case in North Karelia, eastern Finland, is an example of that. They analyse management schemes for mixed stands of Norway spruce, Scots pine and birch, and found that with respect to handling the uncertain growth development under climate change, reactive (anticipatory) approaches are not truly different from a proactive (adaptive) management approach. The reason is that in this specific case, climate change is not likely to drastically impact the health or stability of the ecosystem, but rather it gradually changes the absolute and relative performances of the species. Forest management, however, can capture and react adequately to most of this in a reactive manner. They also find, as is a standard result, that the proactive approach is much more important when it comes to handling price uncertainty.
There may be situations, also in the Boreal areas, where forest managers and society may benefit significantly more from talking a proactive approach. An example of this could be where the choice of tree species to favour regeneration would be better based on expectations than past observations. Or there may be cases, where climate change induced increases in hazard risks may call for early action to avoid losses.
The proactive management approach
The proactive management approach is in the literature also referred to as the fully adaptive management approach. As opposed to the former, the decision maker applying this approach does not only observe the development of the current climate and the state of the forest ecosystems; (s)he also assess likely developments and impacts of climate change using numerous sources of information and observations. The decision maker bases current decisions as much on observed status of the forest as on expectations of future climate change impacts and implication for forest management. If the proactive manager makes good forecasts and forms well-founded expectations, (s)he should perform at least as good as the reactive manager. However, searching for and assessing information is costly and expectations and forecasts may be imprecise, ill-founded or biased, and it is in some situations not obvious that much is gained from such an approach. In other cases, it is more obviously a clear advantage.
In MOTIVE a decision model for even-aged beech stand in Switzerland was developed, assessing the optimal management scheme for a number of site conditions for given future climate developments (Trasobares et al., 2013). However, there is uncertainty with respect to which scenario is the more realistic, so management was optimised (thinning and timing of clear-felling) for 4 different climate change scenarios at various locations. The financial results, in terms of Land expectation values (LEV), from a site with good current growth conditions are shown in Table 1. The diagonal shows the return from the optimal management for the four different climate scenarios. As is seen, two scenarios (A1B_MP1 and A1B_RCA_ECHAM5) will result in an improved return whereas the opposite is true for the more extreme scenario predicted by the AqB_Hadley climate change model. Proactive managers may consider these different scenario optimal strategies and base their decision on their beliefs or expectations about the probability of different scenarios becoming true.
It may be relevant to consider what the consequences are on forest revenue in case we optimize for one climate development, but another is realised. These returns are given by the off diagonal results. As can be seen the losses are up to 8% in case we believe the A1B_MPI climate change scenario to be the one realized and optimize management for that, but it turns out that A1B_Hadley is realised instead. A minimax strategy would therefore suggest that we choose the management optimized for A1B_Hadley, as it has the largest minimum outcome. Optimising the expected return across climate change scenarios requires some beliefs or probabilities of each scenario. If we assume that they are equally likely, i.e. we believe 25% in each, then picking the management strategy optimised for the current climate gives the highest average result (bottom row), though it is only slightly higher than the best alternative. Other locations in Switzerland show other results; in general the poorer the soil, the more severe the consequences of choosing the wrong management strategy.
The results show that the proactive decision approach takes more information into account and the decision maker may gain from this. The specific analyses of Trasobares et al (2013) also demonstrate that while climate change may both bring gains and losses, and while both may be of significant sizes at least in relative terms, the potential losses are not necessarily catastrophic for the current tree generations, even if long term consequences may perhaps be more significant (Hanewinkel et al 2013).
Only a few cases in MOTIVE have considered the issue of tree species choice, which is one of the most far-reaching decisions in forestry where proactive assessment of the potential long term impacts may play the largest role. In one such MOTIVE application (Schou et al., 2012), we analysed the case of a forest manager in Denmark who needs to decide when to harvest an existing, maturing Norway spruce stand and whether to reforest using Norway spruce or oak. Oak is projected to be fairly insensitive to climate change in this region and a likely future main species, whereas this is not so for Norway spruce (Hanewinkel 2012). At current growth conditions Norway spruce economically outperforms oak on most sites, even with mediocre performance. With climate change, increased risks of windthrows, drought-related bark beetle pests and the like may reduce the profitability of Norway spruce, but economic sensitivity analyses reveal that quite a bit of negative impact is needed to reduce the expected economic performance below that of oak.
We investigated the effects on the forest owner’s decision when (s)he faces possible climate change, simplified into three scenarios among which oak is only truly superior to Norway spruce in one, and when (s)he may be uncertain as to which of these scenarios come true for still either 10, 20 or 30 years. Under a proactive decision approach, forthcoming information implies a value of waiting, as can be seen in Table 2 where the expected LEV of making the best decision after the climate change development is evident (€ 2,993) is higher than the expected value of making the best decision before. If we can wait establishing the new stand until we are fairly certain, we may rule out the worst alternative, thereby increasing the expected value. However, if the worst of the modeled climate scenarios does not come true, then it will still be favourable on this location to establish Norway spruce also in the next rotation.
However, waiting has a cost in terms of delaying the harvest of the existing stand. The decision maker needs to trade off this cost against the positive impact of waiting for information evident on the expected value of the new stand. The further into the future the decision maker thinks information will be available, the more costly is waiting, and the more likely (s)he will base her decision on her/his ex ante expectations. Table 3 illustrates the decisions made for different ages of the Norway spruce stand and time periods until the climate change uncertainty is resolved, when (s)he currently find all scenarios equally likely and is risk neutral. As can be seen, with more than 20 years of continued uncertainty, the decision maker will for all ages above 50 years harvest now and reforest with oak. When uncertainty is resolved in 10 years the decision maker postpone harvesting of the 60 year old stands, and wait for certainty.
Again also this case illustrates that proactive decision making and optimal delaying of decisions may outperform other decision approaches. However, the case also shows that if uncertainty is expected to prevail for longer periods, the decision maker may in any case resolve to make decisions based on ex ante lack of information.
The forest owner’s decision and society’s
In many of the cases analysed in MOTIVE, the forest owners may consider the economic consequences of climate change on the current forests health and production likely to be not too severe for the nearest decades. For that reason they may be less likely to engage in dramatic adaptation measures in forest management, and is more likely to resort to reactive decision approaches. However, from a welfare economic perspective, that is society’s point of view; the potential consequences of climate change may be more severe, the reason being that the long-term provision of many ecosystem services like biodiversity conservation, recreational uses, and erosion protection may be more sensitive to climate change. An obvious case is the increased risks of forest fires in Southern Europe, which calls for adaptive management measures that may not be preferable for the individual owners. This potential discrepancy between what adaptive forest management may be optimal for society and what may be optimal for the forest owner should be addressed, It may be wise for society to coordinate the collection of better information on likely impacts and to disseminate and advise forest owners on adaptive measures, or to design policy instruments and regulation measures that create incentives for forest managers to align decisions and management with society’s objectives.
1.8 A decision support system for forest management planning under climate change
José G.Borges Jordi Garcia-Gonzalo, Juan Guerra-Hernandez, Susete Marques and João Palma
Climate change may substantially impact the forest sector around Europe, therefore, forest managers need new tools that aid the efficiency and effectiveness of forest management under changing environmental conditions. Namely, they need decision support systems (DSS) incorporating growth and yield models that are sensitive to environmental changes. A decision support system is a computer-based information system which supports decision making activities. In forest management, a DSS can allow a forest practitioner to evaluate the future consequences of various management decisions. In this chapter, we look at an example of a DSS applied to a eucalypt forest in Portugal.
In the Mediterranean area, several studies point to the warming of winters and to the increase of both the length of the dry season and the frequency of extreme events like forest fires. This will impact growth and survival of plants as well as their geographical distribution and the composition of plant communities. Eucalypt is the most important forest species in Portugal, extending over 812 000 ha corresponding to 26% of the forest territory (according the last national forest inventory). It is the main source of raw material used by the pulp and paper industry, a leading Portuguese export driven industry
In Chamusca, (Central, Portugal), the eucalypt test forest extends over 6138 ha. Chamusca is a rural county 120 km away from Lisbon (Figure 1). The forest landscape was classified into 1722 stands with areas ranging from 4.8 to 18 ha. The current distribution of stand area by age class is very even, with ages ranging from 0 to 16.5 years, and an average age of 8 years. In this case study, a typical eucalyptus rotation may include up to 2 or 3 coppice cuts, each coppice cut being followed by a stool thinning that may leave an average number of shoots per stool ranging from 1 to 2. Harvest ages ranged from 9 to 14 with a 1-year interval. Initial density was 1400 trees per ha.
The Decision Support System
In order to analyse how the forest management should be adapted under changing climatic conditions we developed a DSS (SADfLOR v ecc 1.0). This DSS integrates four independent and compatible modules, encapsulated in one single graphical interface. The four modules are: i) a management information system to store all relevant information about the target forest (e.g. inventory), ii) a prescription generator module including growth and yield functions (GLOB3PG process-based model) iii). a decision module to assemble these alternative prescriptions into a consistent mathematical model which is solved, iv) a reporting module which allows viewing and reporting the results generated by the decision module.
The DSS helps develop an easy formulation of the problem (Figure 2). It provides information to the user in order to develop and adapt forest management plans. The main steps to use in order to evaluate forest management plans under climatic changing conditions are:
1. Selection of study area and inventory data.
2. Selection of the climate change scenarios.
3. Generation of management alternatives.
4. Simulating all management alternatives in all the stands of the forest over the planning horizon (i.e. 30 years).
5. Introducing prices and costs of operations and calculating revenues.
6. Optimizing forest management.
7. Finally, the report with the strategic management plans is produced. The report includes information in tables, maps, word documents.
The proposed approach was first used to assess the climate change effect on the potential eucalypt pulpwood yield and carbon stock in the whole study area over a 30 year time horizon. Results under “current climate” showed that the maximum eucalypt pulpwood yield would be around 2.35 million m3 with a corresponding value of 81.13 million € and total carbon stock was 228.3 Mg C. Under the “climate change” scenario the timber production was reduced to 2.19 million m3, the corresponding forest value was reduced to 74.7 M € and carbon stocks decreased to 212.7 Mg C.
The DSS was further used to check what would happen if the optimal management plans developed for current climate were implemented under climate change conditions. If the forest management plans designed for current climate conditions are not adapted to climate change, the pulpwood yield would be slightly reduced but the harvests in consecutive years would be very uneven.
If the total area cut at different ages is accounted for, results show that under climate change conditions a delay in the optimal year of cuttings is observed. For example, when switching from current climate scenario to the climate change scenario, the area cut when the stands are 9 to 11 years old is reduced by up to 20%, while the area cut when the stands are 12 and 14 years old increases by 12 and 13% (Figure 3). This clearly shows a delay in the harvest ages at landscape level.
Climate changes may substantially impact the forest adding uncertainty to future forest productivity. For this reason forest managers need new tools that may increase the efficiency and the effectiveness of forest management under changing environmental conditions.
The DSS presented here allows the analysis of the impacts of climate change on managed forests. According to the results, the changing environmental conditions will impact forest growth and forest production decisions. In this context, the DSS developed may help forest owners to prepare management plans under uncertain future conditions.
1.9 A web-based ToolBox approach to support adaptive forest management under climate change
Werner Rammer, Christian Schauflinger, Harald Vacik, João HN Palma, Jordi Garcia-Gonzalo, José G Borges, Manfred J. Lexer
The identification and design, selection and implementation of adaptive measures in forest management requires a sound knowledge base as well as tools to support the forest manager. Decision support systems (DSS) are considered as particularly useful for unstructured, ill and semi-structured decision making problems. Thanks to the recent huge advances in information technology, DSS are nowadays usually computer-based tools that help to confront decision makers through direct interaction with data and analysis models. They are, however, usually built for a specific context and specific decision making procedures, making a broader adoption of DSS into the practice difficult.
A typical setting in forest resource management combines one responsible decision maker and a heterogeneous group of stakeholders having a diversity of partly contrasting interests and expectations towards forest management and who are usually not involved formally in decision making processes about forest management. Forest resource planning and decision making deals with highly complex socio-ecological systems with multiple interacting spatial and temporal dimensions. Finding ways and means to communicate findings about forest ecosystems and their management via information technology is a challenge in itself. This is amplified if decision problems include land use and climate change issues, as inherent uncertainty in planning outcomes increases.
Therefore it is likely that a highly pre-determined decision making process (i.e. the decision model) will not be accepted in this area. Moreover, beyond the different procedural approaches to decision making, it is obvious that a single decision support tool will not be sufficient to cover all needs of any kind of decision makers and stakeholders. However, considering that context specificity and flexibility is a key requirement for acceptance of decision support tools by end users calls for a tool box approach where a diverse set of tools is made available to potential users. A tool box approach was also found to be more suitable for addressing different user and problem types simultaneously.
In the context of the MOTIVE project we set out to design and implement a decision support tool box for adaptive forest management which is based on a thorough analysis of contemporary DSS development activities. The objectives of this contribution are to introduce the conceptual frame of the MOTIVE ToolBox decision support system, and to outline the technical implementation of the ToolBox.
The concept of the AFM ToolBox
Based on prior experiences with the development of DSS a number of principles were derived for the design of the MOTIVE AFM ToolBox.
(i) Modularity. The metaphor of a “tool box” hints already at modularity: it should be easy to add new tools (also from third parties) or to exchange existing tools. Similarly, tools should be able to share common elements (e.g. administrative functionalities such as user management, data import and export, saving DSS sessions, printing).
(ii) Internet. Recent technological advances allow the development of web-based decision support tools. Improved internet browsers can run complex web applications which can be accessed with increasing ease due to the widespread availability of broadband internet connections. Specific advantages of a web-based approach are the reduced access barrier (no downloads and installations required) and the adequacy for decision support in a group setting.
(iii) Different types of knowledge and information. The ToolBox should support both interactive, data driven tools and “softer” types of information such as demonstration examples and FAQs. The data for the interactive tools are produced externally using various types of models (e.g. forest ecosystem models, optimization tools).
(iv) Different data sources. The AFM ToolBox offers easy try-out of tools with ready-to-use data from the MOTIVE case study regions. Ultimately, the usefulness of available tools can be efficiently evaluated with data that represent the intended application domain. Users who find a tool useful for their problem domain can then invest in preparing own customized data for use with AFM ToolBox tools.
(v) Problem types. The AFM ToolBox in its current version is designed for (a) the comparative analysis of management alternatives at stand or landscape level with regard to portfolios of ecosystem services which may comprise timber production, carbon sequestration, and nature conservation and biodiversity under current climate and climate change scenarios, (b) the generation of optimized management plans at landscape level. Assessment entities are either stands or a collection of stands (i.e. landscape). The time frame extends up to 100 years. The temporal resolution of the ecosystem service indicators depends on the forest model used.
A) ToolBox overview and data flow
The implementation of the AFM ToolBox consists of a “ToolBox Framework” acting as the “shell” for the data driven tools and the ToolBox website with the knowledge base (see Figure 1). The input data for the tools are stored in the ToolBox database.
External forest models are required to simulate forest development in response to management and climate scenarios and to provide performance indicators of different alternatives either directly as model output, via linker functions establishing a relationship between model output and a suitable ecosystem service indicator, or to feed other specialized models of ecosystem services with forest structure and composition. Such raw data are transferred to the database by a special tool, the AFM ToolBox client (see Figure 1). The client is highly customizable and has the ability to handle the outputs of a diverse set of forest models (e.g. LandClim or PICUS).
B) ToolBox database
The data format of the AFM ToolBox database contains two types of data: on the one hand, it stores the indicators describing the development of the simulated forest stands. The time resolution is annual or lower (e.g. 10 year periods), and the data is on stand and/ or species level (see Table 1). On the other hand, it includes metadata providing context information to the numerical simulation data (see Table 1). As an example, the “site type” is defined using seven attributes (soil type, soil texture, water influence, stoniness, water holding capacity, water supply rating, nutrient supply rating). Attributes are either numerical (e.g. water holding capacity) or use a pre-defined classification scheme (e.g. soil texture is either “sandy”, “loamy”, or “clay”). Subsequently, simulation output data is linked to these metadata types. This approach provides flexibility from the perspective of the data provider and it enables automatic processing of the data by the tools in the ToolBox.
C) Knowledge base
The AFM ToolBox website is the central start page providing access to the AFM knowledge base and the means to start the tools of the ToolBox (Figure 2). The main elements of the knowledge base are a conceptual description of the forest management planning process (Rauscher 1999), a collection of FAQs from the adaptive forest management domain and a set of case study examples from the MOTIVE project. The different types of information are intensively interlinked. In addition, information is tagged based on two archetypic user types (manager/analyst).
The FAQs aim at the most relevant aspects of climate change (How will the future climate look like?), impacts on forests (How may tree species distribution look like under an altered climate?), potential adaptive measures in forest management and silviculture, planning approaches, and how to deal with uncertainty in climate projections. Whenever possible, the knowledge base integrates data from recent research and networking activities (e.g. COST ECHOES on silviculture and forest management for adaptation and mitigation).
The case study examples comprise of several detailed, science based regional examples from all over Europe which were collected in the MOTIVE project. The examples have been prepared so as to share a common structure and are thus easily comparable. The contents cover the regional background and its specific challenges, provide options and recommendations for forest management under climate change, and they give an overview over methodologies and tools that were used in analyzing the case study problem situation.
D) Vulnerability assessment tool
For the AFM ToolBox we have used the vulnerability approach as introduced by Seidl et al. (2011). Sensitivity to impacts and adaptive capacity of forests are placed on a two dimensional surface and are characterized by a set of indicators. The sensitivity indicators represent a set of ecosystem services and are stored in the database for each available management alternative. The indicators for adaptive capacity are qualitative and need to be provided by the user of the tool. For sensitivity indicators the difference between indicator value under baseline climate and the respective value under a climate change scenario is employed in the vulnerability assessment.
The application of the vulnerability assessment tool is split into three general steps: First, the cases for analysis are selected. Secondly, the value-based preferences of the user or user group are defined, and the third step is the interactive analysis of the results.
The cases for analysis can be selected based on the available metadata in the database. For instance, a user may be interested in forest stands that are dominated by beech at sites with a poor water supply. The selection can be further explored using an integrated map or via diagrams. In the second step, the task of the user is to select relevant ecosystem services and assign weights reflecting the relative subjective importance of the indicator/ ecosystem service.
Finally, the results can be visually analyzed using interactive diagrams that provide insight into the perceived impacts of management and climate scenarios and their relation to the adaptive capacity of the forests. Additional analysis diagrams for a detailed analysis of single cases are available.
A special feature of the vulnerability assessment tool is the “group mode”. Here, a facilitator works together with a group of stakeholders on an assessment problem.
E) Optimal Management Plan
The Optimal Management Plan tool (OMP) tool assigns one of the available management options to each stand entity to optimize the objective function at landscape level while meeting constraints (e.g. even flow constraints). This tool is designed to formulate forest management problems in mixed integer programming (MIP) (see Falcão and Borges (2005), Garcia-Gonzalo et al. (2013)). Further it presents a graphical user interface allowing an easy definition of the objective function as well as the constraints.
In contrast to other tools of the ToolBox, the OMP is considered as an expert tool only, because the application of optimization technique per se requires some profound knowledge of the methods. Notwithstanding, the graphical user interface of the tool is designed for easy and visually pleasing use.
The work process is split into four phases: First, the data set for analysis is selected from a list of available data sets in the database. Typically, a data set is a region comprising of forest stands. Second, a specific part of the region (but also the full region) can be specified from a map or a list. Third, the parameters for the optimization process are provided. They consist of an objective function (the variable to maximize or to minimize) as well as flow and target constraints. Additionally, the user can specifies economic parameters such as interest rate or harvesting costs and revenues.
And fourth, the results of an optimization can be viewed and analyzed as a summary, or as more detailed table view. The tool also contains an option to visualize results on top of a Google Maps rendered map.
”Make your own” – Customizing the AFMToolBox
Getting started with the data driven tools of the AFM ToolBox is simple, since it is web based and comes with ample demonstration data from the MOTIVE project. The AFM ToolBox, however, supports also the use of customized data (i.e. data that is generated by the user or for the user). Full control over all aspects of the AFM ToolBox can be exercised with a local installation of the complete AFM ToolBox either directly to the user’s PC or on a local server. This process is facilitated by a download package of the ToolBox containing all necessary underlying software components and the code for the AFM ToolBox.
Technically, the AFM ToolBox builds upon on a number of open source technologies which are frequently used for web development. Since the ToolBox components are open source itself, it can easily be extended or modified by interested parties. Currently, the AFM ToolBox is can be locally installed on Microsoft Windows or Linux platforms.
Discussion and Outlook
Table 2 lists all tools and major functionalities of the MOTIVE AFM ToolBox. It is a balanced set of information, exploration and analysis components. For the AFM ToolBox we have decided to focus on relatively simple graphical representation where the user can shift between several graphical variants to explore effects of climate and management on ecosystem service performance. To promote the idea of an adaptive management approach and to improve the quality of decision making, ample emphasis is on the interlinkage of the tools and the knowledge base.
What are the limitations and the benefits of the current version? Free accessibility via the internet can definitely be seen as a huge advantage in transferring state of the art knowledge and tools to end users.
However, this flexibility easily masks the fact, that the users have not per se access to analyze their own data. The provisioning of own data is technically challenging (operating simulation models and other tools) and very likely beyond the scope of the typical forest manager. In this case, consultancy is indicated to overcome the technical complexity. If, on the other hand, the procedural complexity of a decision support process is high (e.g. group mode of the Vulnerability Assessment tool) a facilitator may be required to fully utilize the potential of the tool.
These two perspectives link back to the initial challenge of implementing decision support systems. We strongly believe that several user profiles need to be considered when developing advanced DSS. For the AFM ToolBox we distinguish the forest manager and the analyst as target users.
Finally, the openness (use of open source and easy extensibility) allows for an extended life time of the ToolBox. Any DSS developer can take up the AFM ToolBox and continue, extend or improve. Future planned developments include new tools (spatially explicit analysis).
As MOTIVE was developing tools to support decision-making in adaptive forest management, it was essential that all relevant stakeholders have been consulted and involved throughout the whole project. A key feature was the development of stakeholder panels in each case study region and on the European level to ensure that the adaptive tools are robust and relevant across a range of different contexts (see 4.1). Key findings of the project have been summarized in various publications (see 4.2) accompanied by a wide range of dissemination and public relation measures (see 4.3). Major scientific publications are listed in chapter 4.4.
Stakeholder and Decision Maker Interactions
A detailed work plan outlining methods of engagement with stakeholder groups throughout the project was developed at the outset in 2009. The stakeholder engagement plan was included in Deliverable Report D7.1 Stakeholder and Decision-Maker Interactions: Annual stakeholder report. Stakeholder interactions linked to other WPs in MOTIVE were coordinated and facilitated with particular focus on the case study regions. The dialogue with stakeholders and potential end users on the MOTIVE decision support tools started in the case study regions to support the design of tools that were compatible with the decision making process in the forestry sector. Collection of stakeholder views on possible management options for dealing with disturbances and handling uncertainties was carried out for case-study areas in Sweden, Portugal and the Netherlands using questionnaires and/or interviews. Some stakeholder workshops were already organized in regional case studies.
Each case study area had a native-language speaker responsible for organizing contacts with the stakeholders and assisting in mapping the stakeholder landscapes, including identification of key stakeholders as well as any hierarchy in decision-making and descriptions of the social and cultural values associated with the forests. To provide more contextual information for the case study profiles, case study representatives provided information on the following topics:
• Understanding the different factors influencing decision making about forest management and planning
• Reflections on the engagement process
The focus of WP7.1 in the final year was to encourage case studies to evaluate the role stakeholder engagement in the development of models, tools and other project outputs and to provide their own perspectives on how successful they feel stakeholder engagement had been during the project. An evaluation of the effectiveness of stakeholder engagement methods helped to identify the extent to which the models and tool developed are likely to be used at the local level as well as lessons learned (important for future stakeholder engagement). The evaluation was divided into five themes:
1.Level of engagement and representation
2.Engagement aims and methods
3.Uptake and use of project outputs
4.Learning and collaboration
5.Influence and decision making
The Decision maker and Stakeholder Advisory Board (DeStAB) was created in the 2nd half of 2010. The first meeting of DeStAB was held on 2 December 2010 in Brussels. Members of DeStAB included regional stakeholder and decision making organisations, EU-wide organisations active in the field of forestry and the European Commission. Learning more about the project increased their interest in the goals and outputs of MOTIVE. The second DeStAB meeting occurred on 30 November – 1 December 2011 in Baden-Baden, Germany. The second meeting concentrated on stakeholder views and perceived options for handling uncertainty and change. The 3rd meeting was held on 9 November 2012 in Catalonia Spain when a final feedback on MOTIVE products was gathered from the stakeholders. Recommendations from the stakeholders included
i)to have better communication of observed climate change impacts on European Forests
ii)to have better understanding of the costs of delaying action.
The meeting was concluded by the participants saying that DeStAB had been very useful for scientists to gain understanding of the perceptions of stakeholders and exchange views and questions. The stakeholders expressed their appreciation for being able to follow MOTIVE over its lifetime and they perceived the exchange of experiences between stakeholders from different regions across Europe as very useful for their work.
Process and synthesize key findings and results of the project
A Policy Brief on Climate Change Impacts and Adaptation in European Forests was finalised in March 2011. The Policy Brief outlined the status of current knowledge on climate change impacts and adaptation in European forests, and identified the challenges that need to be addressed for successful responses in policy and management. The Policy Brief was distributed to the research community, various stakeholders and policy makers, e.g. to members of the EP Committee on Environment, Public Health and Food Safety. A media release for the Policy Brief was published at the MOTIVE website, EFI website, EFI Network News and in AlphaGalileo website (independent source of research news).
Towards the end of the project implementation period, a second EFI Policy Brief “Climate Change in European Forests: How to Adapt” was prepared based on the findings of MOTIVE and published in April 2013 at the Thinkforest event at the European Parliament in Brussels.
It describes various forest management styles in the face of climate change and illustrates how they have been used in different regions of Europe. Regional results presented in this Policy Brief were collected from the MOTIVE case study areas
A mid-term project communication “MOTIVE highlights” was produced in summer 2011. MOTIVE mid-term results originating from each work package and selected case studies were compiled and key findings presented in a news update. The news update was distributed to the research community and various stakeholders, e.g. to members of the Standing Forestry Committee which represents forestry administrations of the EU Member States.
MOTIVE factsheets on European bioclimatic zones and how forestry in different zones can best adapt to climate change were produced in August 2011. The factsheets were developed for educational purposes and were first presented in the Act Now! Forests for Future - ENO conference in September 2011. The conference, run by the global virtual network Environment Online brought together teachers and students from 60 countries around the world. The factsheets are available at the MOTIVE website.
The MOTIVE booklet published in Summer 2013 provides an overview of major research activities and achievements of MOTIVE for a general audience. Topics include: information on climate change projections for Europe with emphasis on implications for forest management; shifts in potential tree species ranges based on climate projections; genetic adaptation of tree species to climate change; mapping of disturbance risks to European forests with focus on wind, fire and bark beetles; decision support tools for adaptive forest management; and the MOTIVE toolbox which aims to allow users to select adaptive management actions and optimize management plans. It also describes the types of decision making approaches and their implications for adapting forest management to climate change. A special section describing the ten case study forests is included. This booklet will give insight into the challenges which climate change presents, and inform readers as to the various potential responses around Europe.
Various MOTIVE related news articles were published through Alphagalileo throughout the project. AlphaGalileo is the world's independent source of research news, and is used by research organisations and media throughout the world. Amongst the published articles were:
• Optimizing forest management under uncertain growth and economic conditions
• Climate change – the future is now: Communicating the need for adaptive management to stakeholders
• Best Practices – Adaptation of European Forests to Climate Change
• New EFI Policy Brief looks at the challenges of adaptive forest management
A series of news articles on MOTIVE results were published in the EFI News in June 2013. These articles gave an overview of the project and presented a few results in more detail, such as i) shifts in tree species habitat suitability following climate change ii) forest owners’ attitudes on adaptive forest management iii) mapping the risk to European forests and iv) lessons learnt for climate change adaptation of the Mediterranean forest. EFI News reaches stakeholders and researches, and is distributed to over 2000 readers.
Dissemination and public relations
A detailed dissemination plan was developed at the start of the project to identify suitable dissemination activities for different target users (D7.2). All material produced during the project had a uniform and easily recognizable visual image. This was facilitated by a project logo and a common image for presentations and other project outputs. A dedicated website was established by PENSOFT providing information about the project. The domain www.motive-project.net was registered for the MOTIVE website and will be available for five years after project end. The website and the Internal Communication Platform (ICP) were created for getting general public acquainted with the project aims and tasks, for presenting the project outputs, and for providing a medium for communication among project participants. The website and the ICP provided the following main functionalities:
1) conventional electronic communication among MOTIVE members
2) long-term archiving of MOTIVE documentation in an internal online library
3) collecting MOTIVE-related data (publications, databases, documentation) in external and internal online libraries
4) promoting MOTIVE results through regular postings of news
5) dissemination of information for forthcoming MOTIVE-related meetings, etc.
From November 2010 to April 2013, MOTIVE website had 16,181 visitors, of which 10,476 were new and 5,705 returning visitors. The MOTIVE website had 45,798 page views. The most visited pages were: Home Page, Deliverables and Publications, News, Document Library. The most visited pages were: Home page, Deliverables and Publications, News, Document Library. The main traffic came from the USA, Germany, Finland, Spain and the United Kingdom.
In addition to original tasks, the discussion forum on Climate Change Impacts and Adaptation in European Forestry was launched. It can still be accessed at the LinkedIn website. This forum connects active professionals working in forest research, forestry and the forest-based sector as well as in policy and decision making and in non-governmental organizations to discuss impacts of climate change and options for adaptation in European forestry. The forum supports stakeholder interaction and dissemination activities. Up to 1200 members have joined the forum since its launch in March 2010.
MOTIVE co-organised a training school on “Adaptation to climate change in forest management” jointly with the COST Action FP0703 ECHOES in September 2011 in Vienna. 4 MOTIVE experts gave lectures including project results targeted towards young researchers (24 PhD students and post-Doc researchers from 19 countries participated in the event).
Organisation of an international open science conference started in the beginning of 2011. The conference was organised together with the European intergovernmental COST Action ECHOES, the European research projects BACCARA and REINFFORCE, the international research projects TRANZFOR and ForEAdapt, and an IUFRO working group on adaptation to climate change. The conference was held on 21-24 May 2012 in Tours, France. Media Centre at the MOTIVE website has a section on the MOTIVE presentations held at the TOURS conference.
During the European Climate Change Adaptation Conference (ECCA), organised in March 2013, MOTIVE co-organised a workshop focusing on the applicability of key forest management adaptation measures in space and time. After presentations by scientists and practitioners, a group discussion addressed the factors that support or impede adaptation in forest management. A list of these factors was compiled for further use. The 35 participants agreed that networking is important in supporting adaptation to climate change. Sharing knowledge about the anticipated changes and what others do about them is a crucial first step to consider one’s own options for adaptation
The ThinkForest seminar "Climate Change in European Forests: How to Adapt" was organised on 25 April 2013 in collaboration with the project MOTIVE. The programme featured presentations by many scientists in MOTIVE. Also MOTIVE case studies were featured as convincing evidence of climate change effects in Catalonia and the practitioner’s view and the effects of climate change in Kronoberg, Sweden were presented. The science-policy discussion underlined the importance of forests in the EU adaptation strategy.
The final seminar organised for the stakeholders and policy makers gathered about 70 participants to the European Parliament in Brussels on 25 April 2013.
List of Websites:
Scientific coordinator: Prof. Dr. Marc Hanewinkel Swiss Federal Research Institute email@example.com Dissemination officer: Dr. Marcus Lindner European Forest Institute firstname.lastname@example.org Project office: Alfons Bieling Forest Research of Baden-Wuerttemberg email@example.com
Grant agreement ID: 226544
1 May 2009
30 April 2013
€ 9 070 933,80
€ 6 961 072,60
FORSTLICHE VERSUCHS- UND FORSCHUNGSANSTALT BADEN-WUERTTEMBERG
Deliverables not available
Grant agreement ID: 226544
1 May 2009
30 April 2013
€ 9 070 933,80
€ 6 961 072,60
FORSTLICHE VERSUCHS- UND FORSCHUNGSANSTALT BADEN-WUERTTEMBERG
Grant agreement ID: 226544
1 May 2009
30 April 2013
€ 9 070 933,80
€ 6 961 072,60
FORSTLICHE VERSUCHS- UND FORSCHUNGSANSTALT BADEN-WUERTTEMBERG