Search Results
Classifying multi-model wheat yield impact response surfaces showing sensitivity to temperature and precipitation change
Crop growth simulation models can differ greatly in their treatment of key processes and hence in their response to environmental conditions. Here, we used an ensemble of 26 process-based wheat models applied at sites across a European transect to compare their sensitivity to changes in temperature (−2 to +9°C) and precipitation (−50 to +50%). Model results were analysed by plotting them as impact response surfaces (IRSs), classifying the IRS patterns of individual model simulations, describing these classes and analysing factors that may explain the major differences in model responses. The model ensemble was used to simulate yields of winter and spring wheat at four sites in Finland, Germany and Spain. Results were plotted as IRSs that show changes in yields relative to the baseline with respect to temperature and precipitation. IRSs of 30-year means and selected extreme years were classified using two approaches describing their pattern. The expert diagnostic approach (EDA) combines two aspects of IRS patterns: location of the maximum yield (nine classes) and strength of the yield response with respect to climate (four classes), resulting in a total of 36 combined classes defined using criteria pre-specified by experts. The statistical diagnostic approach (SDA) groups IRSs by comparing their pattern and magnitude, without attempting to interpret these features. It applies a hierarchical clustering method, grouping response patterns using a distance metric that combines the spatial correlation and Euclidian distance between IRS pairs. The two approaches were used to investigate whether different patterns of yield response could be related to different properties of the crop models, specifically their genealogy, calibration and process description. Although no single model property across a large model ensemble was found to explain the integrated yield response to temperature and precipitation perturbations, the application of the EDA and SDA approaches revealed their capability to distinguish: (i) stronger yield responses to precipitation for winter wheat than spring wheat; (ii) differing strengths of response to climate changes for years with anomalous weather conditions compared to period-average conditions; (iii) the influence of site conditions on yield patterns; (iv) similarities in IRS patterns among models with related genealogy; (v) similarities in IRS patterns for models with simpler process descriptions of root growth and water uptake compared to those with more complex descriptions; and (vi) a closer correspondence of IRS patterns in models using partitioning schemes to represent yield formation than in those using a harvest index. Such results can inform future crop modelling studies that seek to exploit the diversity of multi-model ensembles, by distinguishing ensemble members that span a wide range of responses as well as those that display implausible behaviour or strong mutual similarities.
Balancing trade-offs between ecosystem services in Germany's forests under climate change
Germany's forests provide a variety of ecosystem services. Sustainable forest management aims to optimize the provision of these services at regional level. However, climate change will impact forest ecosystems and subsequently ecosystem services. The objective of this study is to quantify the effects of two alternative management scenarios and climate impacts on forest variables indicative of ecosystem services related to timber, habitat, water, and carbon. The ecosystem services are represented through nine model output variables (timber harvest, above and belowground biomass, net ecosystem production, soil carbon, percolation, nitrogen leaching, deadwood, tree dimension, broadleaf tree proportion) from the process-based forest model 4C. We simulated forest growth, carbon and water cycling until 2045 with 4C set-up for the whole German forest area based on National Forest Inventory data and driven by three management strategies (nature protection, biomass production and a baseline management) and an ensemble of regional climate scenarios (RCP2.6, RCP 4.5, RCP 8.5). We provide results as relative changes compared to the baseline management and observed climate. Forest management measures have the strongest effects on ecosystem services inducing positive or negative changes of up to 40% depending on the ecosystem service in question, whereas climate change only slightly alters ecosystem services averaged over the whole forest area. The ecosystem services 'carbon' and 'timber' benefit from climate change, while 'water' and 'habitat' lose. We detect clear trade-offs between 'timber' and all other ecosystem services, as well as synergies between 'habitat' and 'carbon'. When evaluating all ecosystem services simultaneously, our results reveal certain interrelations between climate and management scenarios. North-eastern and western forest regions are more suitable to provide timber (while minimizing the negative impacts on remaining ecosystem services) whereas southern and central forest regions are more suitable to fulfil 'habitat' and 'carbon' services. The results provide the base for future forest management optimizations at the regional scale in order to maximize ecosystem services and forest ecosystem sustainability at the national scale.
Combining multiple statistical methods to evaluate the performance of process-based vegetation models across three forest stands
Process-based vegetation models are crucial tools to better understand biosphere-atmosphere exchanges and eco-physiological responses to climate change. In this contribution the performance of two global dynamic vegetation models, i.e. CARAIB and ISBACC, and one stand-scale forest model, i.e. 4C, was compared to long-term observed net ecosystem carbon exchange (NEE) time series from eddy covariance monitoring stations at three old-grown European beech (Fagus sylvatica L.) forest stands. Residual analysis, wavelet analysis and singular spectrum analysis were used beside conventional scalar statistical measures to assess model performance with the aim of defining future targets for model improvement. We found that the most important errors for all three models occurred at the edges of the observed NEE distribution and the model errors were correlated with environmental variables on a daily scale. These observations point to possible projection issues under more extreme future climate conditions. Recurrent patterns in the residuals over the course of the year were linked to the approach to simulate phenology and physiological evolution during leaf development and senescence. Substantial model errors occurred on the multi-annual time scale, possibly caused by the lack of inclusion of management actions and disturbances. Other crucial processes defined were the forest structure and the vertical light partitioning through the canopy. Further, model errors were shown not to be transmitted from one time scale to another. We proved that models should be evaluated across multiple sites, preferably using multiple evaluation methods, to identify processes that request reconsideration.
Description and evaluation of the process-based forest model 4C v2.2 at four European forest sites
The process-based model 4C (FORESEE) has been developed over the past 20 years to study climate impacts on forests and is now freely available as an open-source tool. The objective of this paper is to provide a comprehensive description of this 4C version (v2.2) for scientific users of the model and to present an evaluation of 4C at four different forest sites across Europe. The evaluation focuses on forest growth as well as carbon (net ecosystem exchange, gross primary production), water (actual evapotranspiration, soil water content), and heat fluxes (soil temperature) using data from the PROFOUND database. We applied different evaluation metrics and compared the daily, monthly, and annual variability of observed and simulated values. The ability to reproduce forest growth (stem diameter and biomass) differs from site to site and is best for a pine stand in Germany (Peitz, model efficiency ME=0.98). 4C is able to reproduce soil temperature at different depths in Sorø and Hyytiälä with good accuracy (for all soil depths ME > 0.8). The dynamics in simulating carbon and water fluxes are well captured on daily and monthly timescales (0.51 < ME < 0.983) but less so on an annual timescale (ME < 0). This model–data mismatch is possibly due to the accumulation of errors because of processes that are missing or represented in a very general way in 4C but not with enough specific detail to cover strong, site-specific dependencies such as ground vegetation growth. These processes need to be further elaborated to improve the projections of climate change on forests. We conclude that, despite shortcomings, 4C is widely applicable, reliable, and therefore ready to be released to the scientific community to use and further develop the model.
Fire, late frost, nun moth and drought risks in Germany's forests under climate change
Ongoing climate change affects growth and increases biotic and abiotic threats to Germany's forests. We analysed how these risks develop through the mid-century under a variety of climate change scenarios using the process-based forest model 4C. This model allows the calculation of indicators for fire danger, late frost risk for beech and oak, drought stress and nun moth risk. 4C was driven by a set of 4 simulations of future climate generated with the statistical model STARS and with 10 simulations of future climate based on EURO-CORDEX model simulations for the RCP2.6, RCP4.5 and RCP8.5 pathways. A set of about 70000 forest stands (Norway spruce, Scots pine, beech, oak, birch), based on the national forest inventory describing 98.4 % of the forest in Germany, was used together with data from a digital soil map. The changes and the range of changes were analysed by comparing results of a recent time period (1971–2005) and a scenario time period (2011–2045). All indicators showed higher risks for the scenario time period compared to the recent time period, except the late frost risk indicators, if averaged over all climate scenarios. The late frost risk for beech and oaks decreased for the main forest sites. Under recent climate conditions, the highest risk with regard to all five indicators was found to be in the Southwest Uplands and the northern part of Germany. The highest climate-induced uncertainty regarding the indicators for 2011–2045 is projected for the East Central Uplands and Northeast German Plain.
Tree mortality submodels drive simulated long-term forest dynamics: assessing 15 models from the stand to global scale
Models are pivotal for assessing future forest dynamics under the impacts of changing climate and management practices, incorporating representations of tree growth, mortality, and regeneration. Quantitative studies on the importance of mortality submodels are scarce. We evaluated 15 dynamic vegetation models (DVMs) regarding their sensitivity to different formulations of tree mortality under different degrees of climate change. The set of models comprised eight DVMs at the stand scale, three at the landscape scale, and four typically applied at the continental to global scale. Some incorporate empirically derived mortality models, and others are based on experimental data, whereas still others are based on theoretical reasoning. Each DVM was run with at least two alternative mortality submodels. Model behavior was evaluated against empirical time series data, and then, the models were subjected to different scenarios of climate change. Most DVMs matched empirical data quite well, irrespective of the mortality submodel that was used. However, mortality submodels that performed in a very similar manner against past data often led to sharply different trajectories of forest dynamics under future climate change. Most DVMs featured high sensitivity to the mortality submodel, with deviations of basal area and stem numbers on the order of 10–40% per century under current climate and 20–170% under climate change. The sensitivity of a given DVM to scenarios of climate change, however, was typically lower by a factor of two to three. We conclude that (1) mortality is one of the most uncertain processes when it comes to assessing forest response to climate change, and (2) more data and a better process understanding of tree mortality are needed to improve the robustness of simulated future forest dynamics. Our study highlights that comparing several alternative mortality formulations in DVMs provides valuable insights into the effects of process uncertainties on simulated future forest dynamics. © 2019 The Authors.