Abstract Climate change is expected to alter species assemblages by affecting the outcome of competition between species. Investigating processes of competition remains challenging in tree communities, as they unfold over extensive spatio-temporal scales. Here, we used a deep learning-based meta-model trained on 135 million simulated tree responses to climate across Europe to investigate changes in the competitiveness of nine major tree species under future climate. We harnessed projections from local process models to train a Deep Neural Network of forest state transitions to investigate climate-induced changes in competition at continental scale. We found decreasing competitive strength for evergreen conifers across their distribution, while deciduous broadleaved species increased in competitiveness. Most investigated species lost competitive strength at their warm range edges. Consequently, up to 25% of Europe’s forests could experience a change in the dominant tree species until the end of the 21st century, suggesting a profound climate-induced reassembly of Europe’s forests.

Wildfires, insect outbreaks, and storms cause large pulses of tree mortality. Climate change amplifies these forest disturbances, yet their future magnitude and extent remain uncertain. Here, we simulated future forest disturbance regimes at 100-meter resolution across Europe using a deep learning-based simulation framework. Our results show that forest disturbances will continue to increase throughout the 21st century, with disturbed areas more than doubling relative to the recent past under an unabated continuation of climate change. Wildfires are the main agent driving future disturbance change. Changing disturbances result in an increase in young forests, substantially altering Europe's forest demography. Because of their profound implications for forest carbon storage and the habitat value of forest ecosystems, disturbances should be a priority of forest policy and management.

When an even-aged forest stand reaches maturity, it can be renewed within a limited period to maintain the even-aged structure or gradually transformed to an uneven-aged stand. However, there is still debate as to which silvicultural approach is more profitable, conducive to carbon storage, favourable to biodiversity or resilient. In this study, we simulated the evolution of fifteen stands representative of the Walloon forest, whose initial structure was even-aged. The stands were managed according to two contrasting silvicultural approaches (continuation of the even-aged system vs transformation to an uneven-aged one), and the simulations were run with the SSP3-7.0 climate projections produced by five global circulation models. Our simulations indicate that even-aged and uneven-aged silviculture yield similar outcomes in terms of carbon storage. Financial indicators were likewise largely unaffected, except in the oak-beech mixture, where uneven-aged silviculture increased profitability through the substitution of oak by beech. This shift reduced tree species diversity in uneven-aged oak-beech stands. Tree microhabitats, except in beech stands, were more abundant under uneven-aged silviculture. While mean values of forest ecosystem functioning indicators are largely comparable between the two approaches, uneven-aged stands exhibit higher temporal stability. Uneven-aged silviculture produces stands that are more wind-resistant and avoid periods of extreme vulnerability. Our study shows that the most appropriate silviculture may vary depending on the aspect considered. Uneven-aged silviculture has a definite advantage in terms of stability, risk management, and tree microhabitats. However, maintaining even-aged patches at the landscape scale remains important to facilitate the regeneration of shade-intolerant species.

The temporal change of soil chemistry in the forest floor and mineral soil down to a depth of 40 cm was assessed for the 102 permanent plots of the French Network for the Monitoring of Forest Ecosystems (RENECOFOR), over a 15-year period (from 1993–1995 to 2007–2012). In examining the separate and joint evolutions of a large set of parameters, many significant changes were detected reflecting the fact that French forest soils were not in a steady state. A significant increase in soil organic carbon (SOC) stocks was found, mainly in the surface soil (13.0% increase over the forest floor and the 0–10 cm layer). Conversely, the relative increase of the total nitrogen (Ntot) stocks was lower in the surface soil (4.8% increase), and a general and sharp decline of Ntot was detected between 10 and 40 cm depth (12.0% decrease). These results led to a substantial raise of C/N ratio over the whole soil profile. Another major finding is the difference in soil acidification recovery depending on the initial trophic level. In highly acidified contexts (top soil pH H2O < 4.5), increased soil acidification (pH and base saturation decrease, exchangeable Al increase) over the profile was observed while exchangeable base cation (Ca, Mg, K) pools increased. On the other hand, less acidic soils saw their global buffer capacity enhanced. These observations contrast with what is measured in other European inventories. While a previous study carried out on the same plots and over the same period highlighted SOC as a major driver of soil evolution in the top mineral soil, the possible mechanisms behind the large N decrease in the lower mineral soil remain to be confirmed.

Process-based forest models combine biological, physical, and chemical process understanding to simulate forest dynamics as an emergent property of the system. As such, they are valuable tools to investigate the effects of climate change on forest ecosystems. Specifically, they allow testing of hypotheses regarding long-term ecosystem dynamics and provide means to assess the impacts of climate scenarios on future forest development. As a consequence, numerous local-scale simulation studies have been conducted over the past decades to assess the impacts of climate change on forests. These studies apply the best available models tailored to local conditions, parameterized and evaluated by local experts. However, this treasure trove of knowledge on climate change responses remains underexplored to date, as a consistent and harmonized dataset of local model simulations is missing.Here, our objectives were (i) to compile existing local simulations on forest development under climate change in Europe in a common database, (ii) to harmonize them to a common suite of output variables, and (iii) to provide a standardized vector of auxiliary environmental variables for each simulated location to aid subsequent investigations. Our dataset of European stand- and landscape-level forest simulations contains over 1.1 million simulation runs representing 135 million simulation years for more than 13,000 unique locations spread across Europe. The data were harmonized to consistently describe forest development in terms of stand structure (dominant height), composition (dominant species, admixed species), and functioning (leaf area index). Auxiliary variables provided include consistent daily climate information (temperature, precipitation, radiation, vapor pressure deficit) as well as information on local site conditions (soil depth, soil physical properties, soil water holding capacity, plant-available nitrogen). The present dataset facilitates analyses across models and locations, with the aim to better harness the valuable information contained in local simulations for large-scale policy support, and for fostering a deeper understanding of the effects of climate change on forest ecosystems in Europe.

Forests are expected to be strongly affected by modifications in climate and disturbance regimes, threatening their ability to sustain the provision of essential services. Promoting drought-tolerant species or functionally diverse stands have recently emerged as management options to cope with global change. Our study aimed at evaluating the impact of contrasting stand-level management scenarios on the resilience of temperate forests in eastern North America and central-western Europe using the individual process-based model HETEROFOR. We simulated the evolution of eight stands over 100 years under a future extreme climate according to four management scenarios (business as usual- BAU; climate change adaptation- CC; functional diversity approach- FD; no management- NM) while facing multiple disturbances, resulting in a total of 160 simulations. We found that FD demonstrated the greatest resilience regarding transpiration and tree biomass, followed by CC and then BAU, while these three scenarios were equivalent concerning the net primary production. These results were however dependent on forest type: increasing functional diversity was a powerful option to increase the resilience of coniferous plantations whereas no clear differences between BAU and adaptive management scenarios were detected in broadleaved and mixed stands. The FD promoted a higher level of tree species diversity than any other scenario, and all scenarios of management were similar regarding the amount of harvested wood. The NM always showed the lowest resilience, demonstrating that forest management could be an important tool to mitigate adverse effects of global change. Our study highlighted that tree-level process-based models are a relevant tool to identify suitable management options for adapting forests to global change provided that model limitations are considered, and that alternative management options, particularly those based on functional diversity, are promising and should be promoted from now on.

This dataset contains forest floor and mineral soil chemical properties down to 40 cm depth for two soil sampling campaigns carried out at around 15-year time interval (campaign 1 from 1993 to 1995, campaign 2 from 2007 to 2012) within the 102 permanent plots of the French Network for the Monitoring of Forest Ecosystems (RENECOFOR). Data are reported for soil organic carbon (Corg), total nitrogen (Ntot), C/N ratio (CNratio), exchangeable calcium (Caexch), exchangeable magnesium (Mgexch), exchangeable potassium (Kexch), exchangeable aluminium (Alexch), effective cation exchange capacity (ECEC), extractable phosphorus (P), pH CaCl2 and pH H2O.

Process-based forest growth models with spatially explicit representation are relevant tools to investigate innovative silviculture practices and/or climate change effects because they are based on key ecophysiological processes and account for the effects of local competition for resources on tree growth. Such models are rare and are often calibrated for a very limited number of species and rarely for mixed and/or uneven-aged stands, and none are suitable for the temperate forests of Québec. The aim of this study was to calibrate and evaluate HETEROFOR (HETEROgeneous FORest), a process-based and spatially explicit model based on resource sharing, for 23 functionally diverse tree species in forest stands with contrasting species compositions and environmental conditions in southern Quebec. Using data from the forest inventory of Quebec, we evaluated the ability of HETEROFOR to predict the short-term growth (5–16 years) of these species at the tree and stand levels and the long-term dynamics (120 years) of red and sugar maple stands. The comparison between the prediction quality of the calibration and evaluation datasets showed the robustness of the model performance in predicting individual-tree growth. The model reproduced correctly the individual basal area increment (BAI) of the validation dataset, with a mean Pearson's correlation coefficient of 0.56 and a mean bias of 18 %. Our results also highlighted that considering tree position is of importance for predicting individual-tree growth most accurately in complex stands with both vertically and horizontally heterogeneous structures. The model also showed a good ability to reproduce BAI at the stand level, both for monospecific (bias of −3.7 %; Pearson's r=0.55) and multi-species stands (bias of −9.1 %; Pearson's r=0.62). Long-term simulations of red maple and sugar maple showed that HETEROFOR was able to accurately predict the growth (basal area and height) and mortality processes from the seedling stage to the mature stand. Our results suggest that HETEROFOR is a reliable option to simulate forest growth in southern Quebec and to test new forestry practices under future climate scenarios.

Ungulate impacts on forest understory alter tree species composition, with cascading effects on forest functions and resilience against future climate conditions. Indeed, the ungulate browsing pressure on tree seedlings is species-specific and causes contrasted growth reductions that alter tree recruitment rates. Untangling the effects of browsing from the effects of the other factors driving regeneration success is required to guide the forest and ungulate management. In particular, Fagus sylvatica L. strongly dominates temperate Quercus-Fagus forests close to their climax, and it remains unclear if controlling ungulate populations can maintain tree species diversity in naturally regenerated forests. We addressed this question by monitoring 734 pairs of fenced and unfenced 6-m(2) plots across a broad gradient of Cervus elaphus L. abundance in Belgian Quercus-Fagus forests managed by continuous cover forestry. Seedling height, density, and vegetation cover were monitored from 2016 to 2021. Species diversity and ecological affinity for light, temperature, and atmospheric humidity conditions were computed from these measures. With ungulates, the mean growth of Betula pendula Roth. and Sorbus aucuparia L. was negligible, whereas, without ungulates, their growth was higher than the growth of other species. With ungulates, the growth of Fagus sylvatica L. and Picea abies (L.) H. Karst was higher than other species. Quercus (Quercus petreae (Matt.) Liebl and Quercus robur L.) growth was the lowest in all conditions. Finally, Carpinus betulus L. was heavily browsed but still grew higher than its competitors with ungulates. Ungulate browsing can then severely affect seedling growth and likely reduce the diversity of future recruited trees. In the study area, browsing unfavored the regeneration of the species that are less shade tolerant, more-drought tolerant, and more-heat tolerant. It thus accelerates the natural succession and reduces forest resilience to heat and drought events. Such an observation was found valid over a wide study area encompassing contrasting levels of Cervus elaphus L. abundance. Combining further reductions of ungulate populations with foodscape improvement is likely required to maintain species diversity in these forests.

Wild ungulate populations have increased throughout the northern hemisphere in the last decades. Their rising pressure on understory vegetation may be a decisive driver of shifts in forest diversity, with cascading effects on numerous forest functions. In particular, ungulates often thwart climate change adaptation strategies in forests. Ungulate impact has been measured locally and browsing indicators have been developed to monitor changes in ungulate pressure. However, the relationships linking browsing pressure, impact on tree recruitment, and ungulate abundance remain poorly understood. To fill this gap, we conducted a large experiment in 734 pairs of fenced and unfenced 6 m² plots installed across a broad gradient of red deer abundance in oak-beech forests managed with a continuous cover forestry system in Belgium. Height of the dominant seedlings, seedling density, and vegetation cover were monitored yearly in each plot from 2016 to 2021. Species diversity and ecological affinity for light, temperature, and atmospheric humidity conditions were derived from these measures.Ungulates strongly reduced seedling growth, density and cover of understory vegetation. Among the species studied, the early successional species Betula pendula and Sorbus aucuparia were the most affected. These species failed to grow in height when unprotected from ungulates but grew faster than the other species in fenced plots. In contrast, the height growth of late successional species was little or not affected by ungulate browsing. Without protection against browsing, late successional species (Fagus sylvatica, Picea abies) then rapidly dominated the other admixed species. Like the other early successional species, oak seedlings failed to grow significantly in height without protection from browsing. Even when protected, they grew more slowly than all the other species. Oak regeneration will thus rarely succeed without sylvicultural intervention.These observations are clear evidence that ungulates are a key driver of oak-beech forest succession as they cause a bottleneck for the regeneration of early successional species. Ungulates can affect forest succession, reduce species richness, and reduce forest resilience if recruitment of heat- and drought-tolerant species is reduced.Early successional species regeneration can be improved by strongly reducing ungulate abundance, but also probably by temporarily modifying the availability of food resources at landscape level, which would alleviate ungulate pressure and create windows of opportunity for early successional species recruitment. The current cutting regime produces local resource hotspots where ungulate pressure is much higher.

This repository is linked to the paper "Radiative transfer modeling in structurally-complex stands: what aspects matter most?" submitted to Annals of Forest Science and written by Frédéric ANDRÉ (corresponding author), Louis DE WERGIFOSSE, François DE COLIGNY, Nicolas BEUDEZ, Gauthier LIGOT, Vincent GAUTHRAY-GUYÉNET, Benoit COURBAUD and Mathieu JONARD. The repository contains the three following files : CalibrationResults.csv: Bayes factors and summary statistics of parameter estimates for each calibration run ParameterPosteriorDistributions.csv: median values and 90% credible intervals for the parameter posterior distributions StatisticalComparison.csv: statistics (Fractional bias, Root mean square error, Paired Student test, Pearson correlation coefficient, Parameters of the Deming regression between observed and predicted values) used to compare the 'Best model configurations' For more information concerning this repository or the study, please do not hesitate to contact Frédéric ANDRÉ (frederic.andre@uclouvain.be) or Mathieu JONARD (mathieu.jonard@uclouvain.be).

This study aimed to simulate oak and beech forest growth under various scenarios of climate change and to evaluate how the forest response depends on site properties and particularly on stand characteristics using the individual process-based model HETEROFOR. First, this model was evaluated on a wide range of site conditions. We used data from 36 long-term forest monitoring plots to initialize, calibrate, and evaluate HETEROFOR. This evaluation showed that HETEROFOR predicts individual tree radial growth and height increment reasonably well under different growing conditions when evaluated on independent sites. In our simulations under constant CO2 concentration ([CO2]cst) for the 2071-2100 period, climate change induced a moderate net primary production (NPP) gain in continental and mountainous zones and no change in the oceanic zone. The NPP changes were negatively affected by air temperature during the vegetation period and by the annual rainfall decrease. To a lower extent, they were influenced by soil extractable water reserve and stand characteristics. These NPP changes were positively affected by longer vegetation periods and negatively by drought for beech and larger autotrophic respiration costs for oak. For both species, the NPP gain was much larger with rising CO2 concentration ([CO2]var) mainly due to the CO2 fertilisation effect. Even if the species composition and structure had a limited influence on the forest response to climate change, they explained a large part of the NPP variability (44% and 34% for [CO2]cst and [CO2]var, respectively) compared to the climate change scenario (5% and 29%) and the inter-annual climate variability (20% and 16%). This gives the forester the possibility to act on the productivity of broadleaved forests and prepare them for possible adverse effects of climate change by reinforcing their resilience.

The best options to parametrize a radiative transfer model change according to the response variable used for fitting. To predict transmitted radiation, the turbid medium approach performs much better than the porous envelop, especially when accounting for the intra-specific variations in leaf area density but crown shape has limited effects. When fitting with tree growth data, the porous envelop approach combined with the more complex crown shape provides better results. When using a joint optimization with both variables, the better options are the turbid medium and the more detailed approach for describing crown shape and leaf area density. Solar radiation transfer is a key process of tree growth dynamics in forest. Determining the best options to parametrize a forest radiative transfer model in heterogeneous oak and beech stands from Belgium. Calibration and evaluation of a forest radiative transfer module coupled to a spatially explicit tree growth model were repeated for different configuration options (i.e., turbid medium vs porous envelope to calculate light interception by trees, crown shapes of contrasting complexity to account for their asymmetry) and response variables used for fitting (transmitted radiation and/or tree growth data). The turbid medium outperformed the porous envelope approach. The more complex crown shapes enabling to account for crown asymmetry improved performances when including growth data in the calibration. Our results provide insights on the options to select when parametrizing a forest radiative 3D-crown transfer model depending on the research or application objectives.

The change in forest productivity was simulated in six stands in Wallonia (Belgium) following different climate scenarios using a process-based and spatially explicit tree growth model. Simulations revealed a strong and positive impact of the CO 2 fertilization while the negative effect of the transpiration deficit was compensated by longer vegetation periods. The site modulated significantly the forest productivity, mainly through the stand and soil characteristics. Forest net primary production (NPP) reflects forest vitality and is likely to be affected by climate change. Simulating the impact of changing environmental conditions on NPP and two of its main drivers (transpiration deficit and vegetation period) in six Belgian stands and decomposing the site effect. Based on the tree growth model HETEROFOR, simulations were performed for each stand between 2011 and 2100 using three climate scenarios and two CO2 modalities (constant vs time dependent). Then, the climate conditions, soils and stands were interchanged to decompose the site effect in these three components. In a changing climate with constant atmospheric CO2, NPP values remained constant due to a compensation of the negative effect of increased transpiration deficit by a positive impact of longer vegetation periods. With time-dependent atmospheric CO2, NPP substantially increased, especially for the scenarios with higher greenhouse gas (GHG) emissions. For both atmospheric CO2 modalities, the site characteristics modulated the temporal trends and accounted in total for 56 to 73% of the variability. Long-term changes in NPP were primarily driven by CO2 fertilization, reinforced transpiration deficit, longer vegetation periods and the site characteristics.

Tree-ring archives are one of the main sources of information to reconstruct climate variations over the last millennium with annual resolution. The links between tree-ring proxies and climate have usually been estimated using statistical approaches, assuming linear and stationary relationships. Both assumptions may be inadequate, but this issue can be overcome by ecophysiological modelling based on mechanistic understanding. In this respect, the model MAIDEN (Modeling and Analysis In DENdroecology) simulating tree-ring growth from daily temperature and precipitation, considering carbon assimilation and allocation in forest stands, may constitute a valuable tool. However, the lack of local meteorological data and the limited characterization of tree species traits can complicate the calibration and validation of such a complex model, which may hamper palaeoclimate applications. The goal of this study is to test the applicability of the MAIDEN model in a palaeoclimate context using as a test case tree-ring observations covering the 20th century from 21 Eastern Canadian taiga sites and 3 European sites. More specifically, we investigate the model sensitivity to parameter calibration and to the quality of climatic inputs, and we evaluate the model performance using a validation procedure. We also examine the added value of using MAIDEN in palaeoclimate applications compared to a simpler tree-growth model, i.e. VS-Lite. A Bayesian calibration of the most sensitive model parameters provides good results at most of the selected sites with high correlations between simulated and observed tree growth. Although MAIDEN is found to be sensitive to the quality of the climatic inputs, simple bias correction and downscaling techniques of these data improve significantly the performance of the model. The split-sample validation of MAIDEN gives encouraging results but requires long tree ring and meteorological series to give robust results. We also highlight a risk of overfitting in the calibration of model parameters that increases with short series. Finally, MAIDEN has shown higher calibration and validation correlations in most cases compared to VS-Lite. Nevertheless, this latter model turns out to be more stable over calibration and validation periods. Our results provide a protocol for the application of MAIDEN to potentially any site with tree-ring width data in the extratropical region.

Abstract. Given the multiple abiotic and biotic stressors resulting from global changes, management systems and practices must be adapted in order to maintain and reinforce the resilience of forests. Among others, the transformation of monocultures into uneven-aged and mixed stands is an avenue to improve forest resilience. To explore the forest response to these new silvicultural practices under a changing environment, one need models combining a process-based approach with a detailed spatial representation, which is very rare. We therefore decided to develop our own model (HETEROFOR) according to a spatially explicit approach describing individual tree growth based on resource sharing (light, water and nutrients). HETEROFOR was progressively elaborated through the integration of various modules (light interception, phenology, water cycling, photosynthesis and respiration, carbon allocation, mineral nutrition and nutrient cycling) within CAPSIS, a collaborative modelling platform devoted to tree growth and stand dynamics. The advantage of using such a platform is to use common development environment, model execution system, user- interface and visualization tools and to share data structures, objects, methods and libraries. This paper describes the carbon-related processes of HETEROFOR (photosynthesis, respiration, carbon allocation and tree dimensional growth) and evaluates the model performances for a mixed oak and beech stand in Wallonia (Belgium). This first evaluation showed that HETEROFOR predicts well individual radial growth and is able to reproduce size-growth relationships. We also noticed that the more empirical options for describing maintenance respiration and crown extension provide the best results while the process-based approach best performs for photosynthesis. To illustrate how the model can be used to predict climate change impacts on forest ecosystems, the growth dynamics in this stand was simulated according to four IPCC climate scenarios. According to these simulations, the tree growth trends will be governed by the CO2 fertilization effect with the increase in vegetation period length and in water stress also playing a role but offsetting each other.

Climate change affects forest growth in numerous and sometimes opposite ways, and the resulting trend is often difficult to predict for a given site. Integrating and structuring the knowledge gained from the monitoring and experimental studies into process-based models is an interesting approach to predict the response of forest ecosystems to climate change. While the first generation of models operates at stand level, one now needs spatially explicit individual-based approaches in order to account for individual variability, local environment modification and tree adaptive behaviour in mixed and uneven-aged forests that are supposed to be more resilient under stressful conditions. The local environment of a tree is strongly influenced by the neighbouring trees, which modify the resource level through positive and negative interactions with the target tree. Among other things, drought stress and vegetation period length vary with tree size and crown position within the canopy. In this paper, we describe the phenology and water balance modules integrated in the tree growth model HETEROFOR (HETEROgenous FORest) and evaluate them on six heterogeneous sessile oak and European beech stands with different levels of mixing and development stages and installed on various soil types. More precisely, we assess the ability of the model to reproduce key phenological processes (budburst, leaf development, yellowing and fall) as well as water fluxes. Two two-phase models differing regarding their response function to temperature during the chilling period (optimum and sigmoid functions) and a simplified one-phase model are used to predict budburst date. The two-phase model with the optimum function is the least biased (overestimation of 2.46 d), while the one-phase model best accounts for the interannual variability (Pearson's r=0.68). For the leaf development, yellowing and fall, predictions and observations are in accordance. Regarding the water balance module, the predicted throughfall is also in close agreement with the measurements (Pearson's r=0.856; bias =-1.3 %), and the soil water dynamics across the year are well reproduced for all the study sites (Pearson's r was between 0.893 and 0.950, and bias was between −1.81 and −9.33 %). The model also reproduced well the individual transpiration for sessile oak and European beech, with similar performances at the tree and stand scale (Pearson's r of 0.84–0.85 for sessile oak and 0.88–0.89 for European beech). The good results of the model assessment will allow us to use it reliably in projection studies to evaluate the impact of climate change on tree growth in structurally complex stands and test various management strategies to improve forest resilience.

Accurate characterization of forest litter is of high interest for land surface modeling and for interpreting remote sensing observations over forested areas. Due to the large spatial heterogeneity of forest litter, scattering from litter layers has to be considered when sensed using microwave techniques. Here, we apply a full-waveform radar model combined with a surface roughness model to ultrawideband ground-penetrating radar (GPR) data acquired above forest litter during controlled and in situ experiments. For both experiments, the proposed modeling approach successfully described the radar data, with improvements compared to a previous study in which roughness was not directly accounted for. Inversion of the GPR data also provided reliable estimates of the relative dielectric permittivity of the recently fallen litter (OL layer) and of the fragmented litter in partial decomposition (OF layer) with, respectively, averaged values of 1.35 and 3.8 for the controlled experiment and of 3.9 and 7.5 for the in situ experiment. These results show the promising potentialities of GPR for efficient and non-invasive characterization of forest organic layers.

Horizontally stratified media are commonly used to represent naturally occurring and man-made structures, such as soils, roads, and pavements, when probed by ground-penetrating radar (GPR). Electromagnetic (EM) wave scattering from such multilayered media is dependent on the roughness of the interfaces. In this paper, we developed a closed-form asymptotic EM model considering random rough layers based on the scalar Kirchhoff-tangent plane approximation (SKA) model that we combined with planar multilayered media Green's functions. In order to validate our extended SKA model, we conducted simulations using a numerical EM solver based on the finite-difference time-domain (FDTD) method. We modeled a medium with three layers-a base layer of perfect electric conductor (PEC) overlaid by two layers of different materials with rough interfaces. The reflections at the first and at the second interface were both well reproduced by the SKA model for each roughness condition. For the reflection at the PEC surface, the extended SKA model slightly overestimated the reflection, and this overestimation increased with the roughness amplitude. Good agreement was also obtained between the FDTD simulation input values and the inverted root mean square (rms) height estimates of the top interface, while the inverted rms heights of the second interface were slightly overestimated. The accuracy and the performances of our asymptotic forward model demonstrate the promising perspectives for simulating rough multilayered media and, hence, for the full waveform inversion of GPR data to noninvasively characterize soils and materials.