Heterogeneous Terrain Research Articles

Adaptive center constraint for joint release rate estimation and model correction: Multi-scenario validation against wind tunnel experiments

Release rate estimation is crucial for the consequence assessment and emergency decision-making in nuclear accidents. However, inevitable model biases can lead to significant deviations. This study proposes an Adaptive Center Constraint for joint release rate estimation and model correction (ACC joint) for improved robustness and accuracy. It uses a tailored cost function to determine the optimal center constraint, which can automatically adapt to different cases. It was validated against four wind tunnel experiments, which simulated complex dispersion scenarios with densely built-up and highly heterogeneous terrains. The ACC joint method was compared with the Tikhonov and joint correction. The results indicate that the proposed method significantly improves the accuracy of release rate estimation. Compared to the joint correction method, the mean relative error is reduced by 38.6% and 31.4% in the all-measurement and independent validation, respectively. Furthermore, sensitivity analysis reveals that the ACC joint method provides lower mean relative error with different numbers of measurements and shows ultimate stability in all scenarios. It also suggests that measurement sites should be positioned in downwind high-concentration areas and the foot of mountainous areas for reliable estimation. The results from different cost functions verify the scalability of the proposed method, providing potential applications to other complex scenarios.

An automatic method for extracting sand dunes based on slope cost distance from digital elevation models

Sand dunes significantly impact climate, water resources, and human infrastructure, reflecting sand sources and wind conditions. However, their complex and heterogeneous terrain leads to fragmented and poorly continuous dune extraction results. Focusing on the Badain Jaran Desert, this study developed a sand dune extraction method using slope cost distance analysis with FABDEM data. First, positive and negative terrains were segmented using 30 m resolution digital elevation data. Next, slope patches were classified based on specific criteria: slope less than 1.5°, patch area greater than 0.01 km2, and patches entirely within negative terrains. The filtered patches served as source data, and slope raster data were used to calculate slope cost distance. Sand dunes and interdune areas were delineated based on a predefined slope cost threshold. Post-processing, such as smoothing and vector conversion, yielded the final extraction results. The method effectively extracted compound transverse megadunes and pyramid megadunes, addressing misclassification issues and accurately defining interdune areas. It demonstrated faster processing speeds compared to traditional methods, with high precision (89.38 % & 90.37 %), accuracy (96.5 % & 92.89 %), and recall (94.77 % & 98.63 %). The results are more complete and less fragmented, aiding in the extraction of micro-scale dune features and revealing macro-scale spatial patterns, which are beneficial for studying dune formation, development, and wind dynamics.

Data-driven prediction of wind pressure on low-rise buildings in complex heterogeneous terrains

This study presents a data-driven methodology for predicting the pressure coefficient statistics on the windward wall, roof, and leeward wall of low-rise buildings situated downwind of complex heterogeneous terrains. Two types of artificial neural network models were developed: the empirical parameter-based ANN (PANN) and the morphology-based ANN (MANN). Pressure data from wind tunnel tests on the Wind Engineering Research Field Laboratory (WERFL) building model (building height H = 4 m) in complex heterogeneous terrain were used to develop the ANN models. These models were evaluated against a non-linear fitted model to assess their predictive performances. PANN and MANN demonstrated superior performance in capturing the effects of terrain complexity on the mean (Cp,mean) and the root-mean-square (Cp,RMS) wind pressure coefficients for the windward wall, roof, and leeward wall. Optimal prediction was achieved with a terrain patch size of W × L = 4 × 2, equating to a full-scale area of approximately 72 m × 23 m. This suggests that the morphology within approximately 100 m × 50 m (25H× 12.5H) in front of a low-rise building has the greatest correlation with the wind pressure coefficient. Despite lower R2 values for max Cp,RMS on the leeward wall across all models, both PANN and MANN showed promising accuracy for the six outputs studied. Moreover, a global sensitivity analysis confirmed the impact of terrain roughness and complexity on the prediction models particularly on max Cp,RMS, and underscored the dominance of effective roughness length z0,eff and the coefficient of variation of roughness length COVz0 in influencing model outcomes.

Snowmelt and subsurface heterogeneity control tree water sources in a subalpine forest

AbstractIn high mountain areas, snowmelt water is a key—yet fading—hydrological resource, but its importance for soil recharge and tree root water uptake is understudied. In these environments, heterogeneous terrains enhance a highly variable availability of soil and groundwater resources that can be accessed by plants. We conducted a tracer‐based study on a subalpine forest in the Italian Alps. We investigated the isotopic composition (2H and 18O) of snowmelt, precipitation, spring water, soil water—at different locations and depths—and xylem water of twigs taken from alpine larch, Swiss stone pine and alpenrose plants during bi‐weekly field campaigns (growing seasons of 2020 and 2021). Mixing models based on δ18O revealed a large contribution of snowmelt to soil and xylem water, particularly during early summer. We investigated the contribution of water from different soil depths to xylem water, using the sap flow records to date back the end‐member signatures. We found a flexible use of shallow and deeper soil water by the investigated plants, with groundwater more likely used by larger trees and during the late summer. Results based on isotopic data were combined with geophysical observations of the subsurface structure to develop a conceptual model about the different exploitation of water by plants depending on their location (shallow soil on a slope vs. a saturated area). Our study highlights the relevance of snowmelt in high‐elevation terrestrial ecosystems, where heterogeneous substrates shape the water availability at different depths and, in turn, water uptake by plants.

Validation of Doppler Wind Lidar Measurements with an Uncrewed Aircraft System (UAS) in the Daytime Atmospheric Boundary Layer

Abstract One of the most widely used systems for wind speed and direction observations at meteorological sites is based on Doppler wind lidar (DWL) technology. The wind vector derivation strategies of these instruments rely on the assumption of stationary and homogeneous horizontal wind, which is often not the case over heterogeneous terrain. This study focuses on the validation of two DWL systems, operated by the German Weather Service [Deutscher Wetterdienst (DWD)] and installed at the boundary layer field site Falkenberg (Lindenberg, Germany), with respect to measurements from a small, fixed-wing uncrewed aircraft system (UAS) of the type Multi-Purpose Airborne Sensor Carrier (MASC-3). A wind vector intercomparison at an altitude range from 100 to 500 m between DWL and UAS is performed, after a quality control of the aircraft’s data accuracy against a cup anemometer and wind vane mounted on a meteorological mast also operating at the location. Both DWL systems exhibit an overall root-mean-square difference in the wind vector retrieval of less than 22% for wind speed and lower than 18° for wind direction. The enhancement or deterioration of these statistics is analyzed with respect to scanning height and atmospheric stability. The limitations of this type of validation approach are highlighted and accounted for in the analysis.

Deep-learning-derived planetary boundary layer height from conventional meteorological measurements

Abstract. The planetary boundary layer (PBL) height (PBLH) is an important parameter for various meteorological and climate studies. This study presents a multi-structure deep neural network (DNN) model, which can estimate PBLH by integrating the morning temperature profiles and surface meteorological observations. The DNN model is developed by leveraging a rich dataset of PBLH derived from long-standing radiosonde records augmented with high-resolution micro-pulse lidar and Doppler lidar observations. We access the performance of the DNN with an ensemble of 10 members, each featuring distinct hidden-layer structures, which collectively yield a robust 27-year PBLH dataset over the southern Great Plains from 1994 to 2020. The influence of various meteorological factors on PBLH is rigorously analyzed through the importance test. Moreover, the DNN model's accuracy is evaluated against radiosonde observations and juxtaposed with conventional remote sensing methodologies, including Doppler lidar, ceilometer, Raman lidar, and micro-pulse lidar. The DNN model exhibits reliable performance across diverse conditions and demonstrates lower biases relative to remote sensing methods. In addition, the DNN model, originally trained over a plain region, demonstrates remarkable adaptability when applied to the heterogeneous terrains and climates encountered during the GoAmazon (Green Ocean Amazon; tropical rainforest) and CACTI (Cloud, Aerosol, and Complex Terrain Interactions; middle-latitude mountain) campaigns. These findings demonstrate the effectiveness of deep learning models in estimating PBLH, enhancing our understanding of boundary layer processes with implications for improving the representation of PBL in weather forecasting and climate modeling.

Flux‐gradient relations and their dependence on turbulence anisotropy

AbstractMonin–Obukhov similarity theory (MOST) is used in virtually all Earth System Models to parametrize the near‐surface turbulent exchanges and mean variable profiles. Despite its widespread use, there is high uncertainty in the literature about the appropriate parametrizations to use. In addition, MOST has limitations in very stable and unstable regimes, over heterogeneous terrain and complex orography, and has been found to represent the surface fluxes incorrectly. A new approach including turbulence anisotropy as a non‐dimensional scaling parameter has recently been developed and has been shown to overcome these limitations and generalize the flux‐variance relations to complex terrain. In this article, we analyze the flux‐gradient relations for five well‐known datasets, ranging from flat and homogeneous to slightly complex terrain. The scaling relations show substantial scatter and highlight the uncertainty in the choice of parametrization even over canonical conditions. We show that, by including information on turbulence anisotropy as an additional scaling parameter, the original scatter becomes well bounded and new formulations can be developed that drastically improve the accuracy of the flux‐gradient relations for wind shear () in unstable conditions and for temperature gradient () in both unstable and stable regimes. This analysis shows that both and are strongly dependent on turbulence anisotropy and allows us finally to settle the extensively discussed free convection regime for , which clearly exhibits a power law when anisotropy is accounted for. Furthermore, we show that the eddy diffusivities for momentum and heat and the turbulent Prandtl number are strongly dependent on anisotropy and that the latter goes to zero in the free convection limit. These results highlight the necessity to include anisotropy in the study of near‐surface atmospheric turbulence and lead the way for theoretically more robust simulations of the boundary layer over complex terrain.

Influence of different factors on coseismic deformation of the 2015 Mw7.8 earthquake in Nepal

In Geophysics, topographic factors are observations that can be directly measured, but they are often ignored to simplify the model. Studying the coseismic deformation caused by earthquakes helps accurately determine the epicenter's parameterization. It provides a reference for the reasonable layout of coseismic observation stations and GNSS observation stations. After the Mw7.8 earthquake in Nepal in 2015, GCMT, USGS, GFZ, CPPT, and other institutions released their epicenter parameter. However, according to their parameters, the coseismic displacements simulated by the spectral-element method are quite different from the GNSS observations. Firstly, this paper inverts the geometric parameters of the seismogenic fault with Nepal’s coseismic GNSS displacement. The spectral-element method determines the source's location and depth under the heterogeneous terrain and outputs the source parameters. Among the results of many studies, the surface source is more consistent with the generation mechanism of large earthquakes. Secondly, this paper calculates the fault slip distribution of this earthquake using SDM (Steepest Descent Method) based on GNSS and InSAR data, which is divided into 1500 subfaults, and the moment tensor of each subfault is calculated. This paper investigates the distribution characteristics of the coseismic deformation field of the 2015 Mw 7.8 earthquake in Nepal under three different models. The results show that the influence of topographic factors is ~ 20%, and the influence of heterogeneous factors is ~ 10%. This paper concludes that the influence of topographic factors is much more significant than that of heterogeneous factors, and the influence of both should be addressed in coseismic deformation calculations.

Spatiotemporal variability of streamflow in the Pearl River Basin: Controls of land surface processes and atmospheric impacts

AbstractTo advance understanding of the spatiotemporal variability of the streamflow in the Pearl River Basin (PRB), we used the Soil and Water Assessment Tool model to simulate the water fluxes from 2010 to 2020 and to identify the underlying controlling mechanisms. The streamflow in the PRB is highly variable and controlled by complex land surface processes and atmospheric impacts over the heterogeneous terrain. Two key factors primarily govern the streamflow: (1) the location of the active precipitation zone, which is determined by the interaction between the monsoon path and uplifting effect of the terrain, and (2) the redistributions resulting from land use and soil characteristics. We observe distinct patterns in the different water fluxes across the different regions. Specifically, surface flow exhibits the highest activity within the precipitation zone. Lateral flow and actual evapotranspiration (AET) have the greatest intensity in the forests and in agricultural regions, respectively, and the aquifer flow is more active in areas with coarse soil textures. The land surface processes of the AET and aquifer retention significantly govern the temporal variability of the streamflow, contributing to the precipitation and streamflow being out of phase in the PRB. Based on the underlying mechanisms driving streamflow variability, we classify the PRB into three substreams: a drought‐prone upstream, a hydrologically active midstream, and a typhoon‐affected downstream, and each substream exhibits distinct spatiotemporal characteristics of streamflow. We find that the time series and probability distributions of the streamflow at different tributaries within each substream are similar. Each probability distribution is multimodal and can be decomposed into three unimodal distributions representing dry, transitional, and wet conditions. Specifically, the PRB features a large and steep dry mode, a flat transitional mode, and a short wet mode.

Probabilistic slope stability analysis: A novel distribution for soils exhibiting highly variable spatial properties

Slope stability calculation depends on the soil properties (cohesion and the friction angle) of the soil. Heterogeneous terrains are frequently observed in civil and mining projects where the properties are highly spatially variable. Based on a real data from case studies, this paper presents a probabilistic analysis of the slope stability of highly heterogeneous terrains with a very high coefficient of variation (COV) of the cohesion distribution. The existing deterministic and probabilistic approaches for calculating slope stability lack the capability to effectively consider the significant heterogeneity present in the terrain The objective of the paper is to develop a new bounded interval distribution having a COV that is as high (>150%) as the COV of the cohesion distribution The results obtained with this new distribution are compared to 4 other semi-infinite distributions. To consider the correlation between cohesion and the friction angle, a specific formulation was developed to generate friction angles varying between fixed minimum and maximum limits and having the desired correlation coefficient, mean, and standard deviation. The new cohesion and friction angle distributions were incorporated and tested in a probabilistic numerical model. The new distribution can presently be applied to geotechnical studies for terrains and heterogenous materials with properties exhibiting high spatial variability.

Geotechnical characterisation of coal spoil piles using high-resolution optical and multispectral data: A machine learning approach

Geotechnical characterisation of spoil piles has traditionally relied on the expertise of field specialists, which can be both hazardous and time-consuming. Although unmanned aerial vehicles (UAV) show promise as a remote sensing tool in various applications; accurately segmenting and classifying very high-resolution remote sensing images of heterogeneous terrains, such as mining spoil piles with irregular morphologies, presents significant challenges. The proposed method adopts a robust approach that combines morphology-based segmentation, as well as spectral, textural, structural, and statistical feature extraction techniques to overcome the difficulties associated with spoil pile characterisation. Additionally, it incorporates minimum redundancy maximum relevance (mRMR) based feature selection and machine learning-based classification. This automated characterisation will serve as a proactive tool for dump stability assessment, providing crucial data for improved stability models and contributing to a greener and more responsible mining industry .

A comparative analysis of the applicability of typical land cover datasets in the karst strongly heterogeneous terrain in Southern China

The international scientific community pays close attention to global land surface remote sensing mapping. Accurate LULC datasets are crucial for sustainable development assessments at various scales. Taking the southern China’s karst as an example, this study selected three sets CRLC-10m (CRLC Land Cover), ESRI-10m (ESRI Land Cover), and ESA-10m (ESA World Cover) were select for 2020 to conduct correlation analysis, spatial consistency analysis and accuracy evaluation. The results indicate that: (1) There is good correlation between ESRI-10m and ESA-10m; (2) Only 3.67% of the study area shows full consistency across all three datasets, with good spatial consistency in crop, forest, grassland types; (3) The overall accuracy of the land cover data sets in the study area is CRLC-10m (55.40%), ESRI-10m (55.26%), and ESA-10m (49.40%), and the accuracy of CRLC-10m is less affected by slope changes. This study provides theoretical guidance for selecting LULC data and enhancing remote sensing classification accuracy in karst regions.

Experimental investigation on influence of terrain complexity for wind pressure of low-rise building

This study conducted extensive wind tunnel tests to evaluate the impact of terrain complexity on wind pressure across low-rise buildings. A series of wind tunnel tests was performed using 50 actual terrain morphologies in the US. The findings were compared with results obtained from testing on homogeneous terrain to discern variations in pressure coefficients. A notable increase in turbulence intensity was observed in complex heterogeneous terrains, even with similar effective roughness lengths. The heightened turbulence property was a crucial factor in explaining changes in Cp,mean. The magnitude of Cp,mean demonstrated a continuous rise in the windward wall and roof 1 regions with increasing turbulence intensity. This correlation held true even for Iu,eave values surpassing 0.3. In contrast, while Cp,RMS exhibited a tendency to increase with rising Iu,eave, it did not exhibit the same continuous increase phenomenon. Consequently, no significant disparity in magnitude was noted between homogeneous and heterogeneous terrains in this regard. These findings underscore the importance of accounting for terrain complexity in wind load assessments, particularly in scenarios of heightened terrain diversity, to mitigate potential errors in wind load evaluations.

Mechanical intelligence simplifies control in terrestrial limbless locomotion.

Limbless locomotors, from microscopic worms to macroscopic snakes, traverse complex, heterogeneous natural environments typically using undulatory body wave propagation. Theoretical and robophysical models typically emphasize body kinematics and active neural/electronic control. However, we contend that because such approaches often neglect the role of passive, mechanically controlled processes (those involving "mechanical intelligence"), they fail to reproduce the performance of even the simplest organisms. To uncover principles of how mechanical intelligence aids limbless locomotion in heterogeneous terradynamic regimes, here we conduct a comparative study of locomotion in a model of heterogeneous terrain (lattices of rigid posts). We used a model biological system, the highly studied nematode worm Caenorhabditis elegans, and a robophysical device whose bilateral actuator morphology models that of limbless organisms across scales. The robot's kinematics quantitatively reproduced the performance of the nematodes with purely open-loop control; mechanical intelligence simplified control of obstacle navigation and exploitation by reducing the need for active sensing and feedback. An active behavior observed in C. elegans, undulatory wave reversal upon head collisions, robustified locomotion via exploitation of the systems' mechanical intelligence. Our study provides insights into how neurally simple limbless organisms like nematodes can leverage mechanical intelligence via appropriately tuned bilateral actuation to locomote in complex environments. These principles likely apply to neurally more sophisticated organisms and also provide a design and control paradigm for limbless robots for applications like search and rescue and planetary exploration.

Polarimetric Synthetic Aperture Radar Image Classification Based on Double-Channel Convolution Network and Edge-Preserving Markov Random Field

Deep learning methods have gained significant popularity in the field of polarimetric synthetic aperture radar (PolSAR) image classification. These methods aim to extract high-level semantic features from the original PolSAR data to learn the polarimetric information. However, using only original data, these methods cannot learn multiple scattering features and complex structures for extremely heterogeneous terrain objects. In addition, deep learning methods always cause edge confusion due to the high-level features. To overcome these limitations, we propose a novel approach that combines a new double-channel convolutional neural network (CNN) with an edge-preserving Markov random field (MRF) model for PolSAR image classification, abbreviated to “DCCNN-MRF”. Firstly, a double-channel convolution network (DCCNN) is developed to combine complex matrix data and multiple scattering features. The DCCNN consists of two subnetworks: a Wishart-based complex matrix network and a multi-feature network. The Wishart-based complex matrix network focuses on learning the statistical characteristics and channel correlation, and the multi-feature network is designed to learn high-level semantic features well. Then, a unified network framework is designed to fuse two kinds of weighted features in order to enhance advantageous features and reduce redundant ones. Finally, an edge-preserving MRF model is integrated with the DCCNN network. In the MRF model, a sketch map-based edge energy function is designed by defining an adaptive weighted neighborhood for edge pixels. Experiments were conducted on four real PolSAR datasets with different sensors and bands. The experimental results demonstrate the effectiveness of the proposed DCCNN-MRF method.

Comparison of Hydrodynamic Processes Modelling Results in Shallow Water Bodies Based on 3D Model and 2D Model Averaged by Depth

Introduction. Two-dimensional hydrodynamic models have proven their ability to adequately describe the processes of runoff and transportation in rivers, lakes, estuaries, deltas and seas. Practice shows that even where significant three-dimensional effects are expected, for example, with wind flows, a two-dimensional approach can work effectively. However, in some cases, the two-dimensional model does not accurately reflect the actual flow structures. For example, in shallow waters with complex bathymetry, heterogeneous terrain and dynamics can lead to a non-uniform velocity profile. The aim of the study is to develop a basis for determining in which cases a two-dimensional model averaged in depth is sufficient for modelling hydrodynamic processes in shallow waters like the Azov Sea, and in which cases it is advisable to use a three-dimensional model to obtain accurate results.Materials and Methods. Local analytical solutions have been obtained for the propagation of the predominant singular progressive wave in a shallow, well-mixed reservoir. Advective terms and Coriolis terms are neglected, the vortex viscosity is assumed to be constant, and the lower friction term is linearized. Special attention is paid to the latter, since the characteristics of the models significantly depend on the method of determining the coefficients of lower friction. The analytical method developed in the study shows that certain combinations of higher flow velocities (u ≈˃ 1 m/s) and water depths (d ˃ 50 m) can cause significant differences between the results of the depth-averaged model and the model containing vertical information.Results. The results obtained are verified by numerical simulation of stationary and non-stationary periodic flows in a schematized rectangular basin. The results obtained as a result of three-dimensional modelling are compared with the results of two-dimensional modelling averaged in depth. Both simulations show good compliance with analytical solutions.Discussion and Conclusions. Analytical solutions were found by linearization of the equations, which obviously has its limitations. A distinction is made between two types of nonlinear effects — nonlinearities caused by higher-order terms in the equations of motion, i.e. terms of advective acceleration and friction, and nonlinear effects caused by geometric nonlinearities, this is due, for example, to different water depths and reservoir widths, which will be important when modelling a real sea.

Experimental study on wind characteristics and prediction of mean wind profile over complex heterogeneous terrain

This study investigates the impact of complex heterogeneous terrain on mean wind speed and turbulence intensity, highlighting the significance of terrain configuration in the wind loading on buildings and airflow over urban areas. Extensive wind tunnel tests were conducted for 60 different roughness configurations, obtained by processing aerial images across the United States. The study makes two main contributions. First, a model was proposed to predict mean wind profiles, using the morphological information of complex heterogeneous terrain. The Deaves and Harris model was utilized along with a novel algorithm for the automatic characterization of roughness transitions. The proposed model exhibited less than 2% average prediction error compared to the measured wind speed. Second, the study investigated the impact of terrain complexity on near-surface wind characteristics. By comparing the experimental results with those obtained from a homogeneous terrain with a similar roughness length, we quantified the potential errors that may arise when assuming a homogeneous terrain for wind speed assessment. It was observed that increasing variability in roughness length led to a decrease in mean wind speed and an increase in turbulence intensity. The influence of terrain complexity, however, was found to be secondary compared to roughness length. Consequently, the relationships between terrain complexity and wind characteristics were quantified, and a simplified model was proposed.

Three‐Dimensional Differentiation of the Contribution of Climatic Factors to Vegetation Change in the Pan‐Tibetan Plateau

AbstractDue to the heterogeneous terrains and unique alpine climate environment, the three‐dimensional differentiation of climatic impacts on vegetation change and its possible mechanism in the Pan‐Tibetan Plateau (PTP) are complex and lack adequate research. Hence, we here quantify the three‐dimensional differentiation of the contribution of climatic factors to vegetation change using the analysis of variance (ANOVA) method. Results indicate that the overall PTP has been warmer and wetter but partially drier, with a remarkable fractional vegetation coverage (FVC) increase since 2000. Moreover, the total areas with FVC changes accounting for approximately 49.3% of PTP are significantly driven by climate factors. Precipitation and potential evaporation predominate the vegetation change at relatively high altitudes and latitudes, while temperature affects the vegetation change at relatively low latitudes and parts of low elevations. Meanwhile, relative humidity dominates the vegetation change at middle altitudes and latitudes. Furthermore, the hydrothermal conditions and climate change magnitudes modulate the three‐dimensional differentiation. On the one hand, the effects of heat conditions on FVC change overwhelm the moisture conditions in alpine areas or warm and wet regions, in sharp contrast with the dominant role of water conditions in most dry areas. On the other hand, the spatial heterogeneity in climate change magnitudes influences the contribution of climatic factors in most hydrothermal conditions. Additionally, interventions challenging to quantify, such as climatic and mountainous hazards, anthropogenic activities, and vegetation feedback, will lead to the differentiation’s uncertainties. Our research hopefully provides theoretical support for national ecosystem management and major construction projects across PTP.

Adaptation of QES-Fire, a dynamically coupled fast response wildfire model for heterogeneous environments

Background Modelling of fire front progression is challenging due to the large range of spatial and temporal scales involved in the interactions between the atmosphere and fire fronts. Further modelling complications arise when heterogeneous terrain and fuels are considered. Aims The aim of this study was to create a new parameterisation for wildfire-induced winds that accounts for the effects of heterogeneous terrain and fuels within the QES-Fire modelling framework – a fast-response wildfire model. Methods QES-Fire’s new turbulent plume merging model allows for distinct plumes to be merged together from fires burning in heterogeneous terrain with heterogeneous fuels. Additionally, fuel inputs from the LANDFIRE database developed for the Rothermel rate of spread (ROS) model, are translated to the Balbi ROS model. Key results The model was evaluated against the forested RxCADRE field experiment, with and without the effects of heterogeneity. Inclusion of heterogeneity reduced the relative error in burned area from 36 to 6%. Conclusions Small variations in terrain and fuel heterogeneity lead to large errors in rate and direction of fire front spread. Implications The modelled effects of terrain and fuel heterogeneity indicated the importance of capturing the complex coupled wildfire–atmospheric dynamics at the fire front.

Topography alters stand structure, carbon stocks and understorey species composition of Cedrela odorata plantation, in a semi-deciduous forest zone, Ghana

Pure plantations established across different topographic positions may respond differently and exhibit varied stand characteristics. However, there is limited information on variability of pure plantation stand attributes along topographic gradients, particularly in West Africa. This study examined the stand structure, carbon stocks and understorey woody species composition of relatively older Cedrela plantations on three topographic positions within the deciduous forest zone of Ghana. About 15, 40 × 40 m plots were surveyed in 23-year-old Cedrela odorata plantations located on hill top, flat terrain and valley bottoms in the Tinte Bepo Forest Reserve, Ghana. Data were analysed using ANCOVA, Bayesian modelling of diameter distribution and correspondence analysis.Diameter at breast height (DBH; p<0.0001), total height (p<0.025), and stem density (p<0.001) were significantly different among the three topographic areas. Trees on the valley bottom had the highest DBH, total height, and biomass carbon while the highest stem density and basal area occurred on the flat terrain. Slope explained 63% of the total variability in the stem density of the C. odorata stand. Bayesian modelling of diameter distribution revealed that the 3-parameter Weibull function better fitted the diameter distribution of the trees on the hill top while the Johnson SB was more appropriate at the flat terrain and the valley bottom. Understorey woody species composition was significantly different among geomorphic positions. Differences in aboveground biomass carbon stocks among topographic positions were not significant.The local topography therefore severely alters plantation stand structure, leading to different distribution models being adopted at different topographic positions. Furthermore, cedrela plantations facilitate climate mitigation, species recruitment and functional/trait diversity differences along topographic gradients. This has implications for large scale plantation development especially on heterogeneous terrains.