Comparison of sensitivity analysis techniques: A case study with the rice model WARM

The considerable complexity often included in biophysical models leads to the need of specifying a large number of parameters and inputs, which are available with various levels of uncertainty. Also, models may behave counter-intuitively, particularly when there are nonlinearities in multiple input-output relationships. Quantitative knowledge of the sensitivity of models to changes in their parameters is hence a prerequisite for operational use of models. This can be achieved using sensitivity analysis (SA) via methods which differ for specific characteristics, including computational resources required to perform the analysis. Running SA on biophysical models across several contexts requires flexible and computationally efficient SA approaches, which must be able to account also for possible interactions among parameters. A number of SA experiments were performed on a crop model for the simulation of rice growth (Water Accounting Rice Model, WARM) in Northern Italy. SAs were carried out using the Morris method, three regression-based methods (Latin hypercube sampling, random and Quasi-Random, LpTau), and two methods based on variance decomposition: Extended Fourier Amplitude Sensitivity Test (E-FAST) and Sobol’, with the latter adopted as benchmark. Aboveground biomass at physiological maturity was selected as reference output to facilitate the comparison of alternative SA methods. Rankings of crop parameters (from the most to the least relevant) were generated according to sensitivity experiments using different SA methods and alternate parameterizations for each method, and calculating the top-down coefficient of concordance (TDCC) as measure of agreement between rankings. With few exceptions, significant TDCC values were obtained both for different parameterizations within each method and for the comparison of each method to the Sobol’ one. The substantial stability observed in the rankings seem to indicate that, for a crop model of average complexity such as WARM, resource intensive SA methods could not be needed to identify most relevant parameters. In fact, the simplest among the SA methods used (i.e., Morris method) produced results comparable to those obtained by methods more computationally expensive.     

Using global sensitivity analysis of demographic models for ecological impact assessment

Population viability analysis (PVA) is widely used to assess population‐level impacts of environmental changes on species. When combined with sensitivity analysis, PVA yields insights into the effects of parameter and model structure uncertainty. This helps researchers prioritize efforts for further data collection so that model improvements are efficient and helps managers prioritize conservation and management actions. Usually, sensitivity is analyzed by varying one input parameter at a time and observing the influence that variation has over model outcomes. This approach does not account for interactions among parameters. Global sensitivity analysis (GSA) overcomes this limitation by varying several model inputs simultaneously. Then, regression techniques allow measuring the importance of input‐parameter uncertainties. In many conservation applications, the goal of demographic modeling is to assess how different scenarios of impact or management cause changes in a population. This is challenging because the uncertainty of input‐parameter values can be confounded with the effect of impacts and management actions. We developed a GSA method that separates model outcome uncertainty resulting from parameter uncertainty from that resulting from projected ecological impacts or simulated management actions, effectively separating the 2 main questions that sensitivity analysis asks. We applied this method to assess the effects of predicted sea‐level rise on Snowy Plover (Charadrius nivosus). A relatively small number of replicate models (approximately 100) resulted in consistent measures of variable importance when not trying to separate the effects of ecological impacts from parameter uncertainty. However, many more replicate models (approximately 500) were required to separate these effects. These differences are important to consider when using demographic models to estimate ecological impacts of management actions.     

Application of a three-dimensional model for assessing effects of small clear-cuttings on radiation and soil temperature

A three-dimensional model Mixfor-3D of soil–vegetation–atmosphere transfer (SVAT) was developed and applied to estimate possible effects of tree clear-cutting on radiation and soil temperature regimes of a forest ecosystem. The Mixfor-3D model consists of several closely coupled 3D sub-models describing: forest stand structure; radiative transfer in a forest canopy; turbulent transfer of sensible heat, H₂O and CO₂ between ground surface and the atmospheric surface layer; evapotranspiration of ground surface vegetation and soil; heat and moisture transfer in soil. The model operates with the horizontal grid resolution, 2 m × 2 m; vertical resolution, 1 m and primary time step, 1 h.     

Modelling the impacts of land-use and drainage density on the water balance of a lowland–floodplain landscape in northeast Germany

This study presents the modelling approach and impact assessment of different strategies for managing wetland water resources and groundwater dynamics of landscapes which are characterised by the hydrological interactions of floodplains and the adjacent lowlands. The assessment of such impacts is based on the analysis of simulation results of complex scenarios of land-use changes and changes of the density of the drainage-network. The method has been applied to the 198 km² Lower Havel River catchment as a typical example of a lowland–floodplain landscape. The model used consists of a coupled soil water and groundwater model, where the latter one is additionally coupled to the surface channel network. Thus, the hydrological processes of the variable saturated soil zone as well as lateral groundwater flow and the interactions between surface water and groundwater are simulated in an integrated manner. The model was validated for several years of significantly different meteorological conditions. The comparison of lateral and vertical water balance components showed the dominance of lateral flow processes and the importance of the interactions between surface water and groundwater for the overall water balance and the hydrological state of that type of landscape.The simulation of land-use change scenarios showed only minor effects of land-use change on the water balance and groundwater recharge. Changes of groundwater recharge were particularly small within the wetland areas being part of the floodplain where interactions between surface water and groundwater are most pronounced. Alterations in vertical groundwater recharge were counter-balanced by the lateral interaction between groundwater and surface water. More significant deviations in groundwater recharge and storage were observed in the more peripheral areas towards the catchment boundaries which are characterised by greater groundwater distance from the surface and less intense of ground water–surface water interactions.However, the simulation results assuming a coarsening of the drainage network density showed the importance of drainage structure and geometry for the water balance: The removal of the artificial draining ditches in the floodplain would result in significant alterations of total groundwater recharge, i.e., less recharge from winter to early summer and an increase of groundwater recharge during summer and autumn. Furthermore the different effects of groundwater recharge alterations on the dynamics of groundwater stages within the wetland areas close to the floodplains compared to the more peripheral areas could be quantified. Finally, it will be discussed that a well-adjusted co-ordination of different management measures is required to reach a sustainable water resources management of such lowland–floodplain landscapes.     

SOMPROF: A vertically explicit soil organic matter model

Most current soil organic matter (SOM) models represent the soil as a bulk without specification of the vertical distribution of SOM in the soil profile. However, the vertical SOM profile may be of great importance for soil carbon cycling, both on short (hours to years) time scale, due to interactions with the soil temperature and moisture profile, as well as on long (years to centuries) time scale because of depth-specific stabilization mechanisms of organic matter. It is likely that a representation of the SOM profile and surface organic layers in SOM models can improve predictions of the response of land surface fluxes to climate and environmental variability. Although models capable of simulating the vertical SOM profile exist, these were generally not developed for large scale predictive simulations and do not adequately represent surface organic horizons. We present SOMPROF, a vertically explicit SOM model, designed for implementation into large scale ecosystem and land surface models. The model dynamically simulates the vertical SOM profile and organic layer stocks based on mechanistic representations of bioturbation, liquid phase transport of organic matter, and vertical distribution of root litter input. We tested the model based on data from an old growth deciduous forest (Hainich) in Germany, and performed a sensitivity analysis of the transport parameters, and the effects of the vertical SOM distribution on temporal variation of heterotrophic respiration. Model results compare well with measured organic carbon profiles and stocks. SOMPROF is able to simulate a wide range of SOM profiles, using parameter values that are realistic compared to those found in previous studies. Results of the sensitivity analysis show that the vertical SOM distribution strongly affects temporal variation of heterotrophic respiration due to interactions with the soil temperature and moisture profile.     

Optimization of ecosystem model parameters using spatio-temporal soil moisture information

Parameters in process-based terrestrial ecosystem models are often nonlinearly related to the water flux to the atmosphere, and they also change temporally and spatially. Therefore, for estimating soil moisture, process-based terrestrial ecosystem models inevitably need to specify spatially and temporally variant model parameters. This study presents a two-stage data assimilation scheme (TSDA) to spatially and temporally optimize some key parameters of an ecosystem model which are closely related to soil moisture. At the first stage, a simplified ecosystem model, namely the Boreal Ecosystem Productivity Simulator (BEPS), is used to obtain the prior estimation of daily soil moisture. After the spatial distribution of 0–10 cm surface soil moisture is derived from remote sensing, an Ensemble Kalman Filter is used to minimize the difference between the remote sensing model results, through optimizing some model parameters spatially. At the second stage, BEPS is reinitialized using the optimized parameters to provide the updated model predictions of daily soil moisture. TSDA has been applied to an arid and semi-arid area of northwest China, and the performance of the model for estimating daily 0–10 cm soil moisture after parameter optimization was validated using field measurements. Results indicate that the TSDA developed in this study is robust and efficient in both temporal and spatial model parameter optimization. After performing the optimization, the correlation (r²) between model-predicted 0–10 cm soil moisture and field measurement increased from 0.66 to 0.75. It is demonstrated that spatial and temporal optimization of ecosystem model parameters can not only improve the model prediction of daily soil moisture but also help to understand the spatial and temporal variation of some key parameters in an ecosystem model and the corresponding ecological mechanisms controlling the variation.     

A model of surface fire,climate and forest pattern in the Sierra Nevada,California

A spatially explicit forest gap model was developed for the Sierra Nevada, California, and is the first of its kind because it integrates climate, fire and forest pattern. The model simulates a forest stand as a grid of 15×15 m forest plots and simulates the growth of individual trees within each plot. Fuel inputs are generated from each individual tree according to tree size and species. Fuel moisture varies both temporally and spatially with the local site water balance and forest condition, thus linking climate with the fire regime. Fires occur as a function of the simulated fuel loads and fuel moisture, and the burnable area is simulated as a result of the spatially heterogeneous fuel bed conditions. We demonstrate the model’s ability to couple the fire regime to both climate and forest pattern. In addition, we use the model to investigate the importance of climate and forest pattern as controls on the fire regime. Comparison of model results with independent data indicate that the model performs well in several areas. Patterns of fuel accumulation, climatic control of fire frequency and the influence of fuel loads on the spatial extent of fires in the model are particularly well-supported by data. This model can be used to examine the complex interactions among climate, fire and forest pattern across a wide range of environmental conditions and vegetation types. Our results suggest that, in the Sierra Nevada, fuel moisture can exert an important control on fire frequency and this control is especially pronounced at sites where most of the annual precipitation is in the form of snow. Fuel loads, on the other hand, may limit the spatial extent of fire, especially at elevations below 1500 m. Above this elevation, fuel moisture may play an increasingly important role in limiting the area burned.     

The MONICA model: Testing predictability for crop growth, soil moisture and nitrogen dynamics

A fundamentally revised version of the HERMES agro-ecosystem model, released under the name of MONICA, was calibrated and tested to predict crop growth, soil moisture and nitrogen dynamics for various experimental crop rotations across Germany, including major cereals, sugar beet and maize. The calibration procedure also included crops grown experimentally under elevated atmospheric CO₂ concentration. The calibrated MONICA simulations yielded a median normalised mean absolute error (nMAE) of 0.20 across all observed target variables (n = 42) and a median Willmott's Index of Agreement (d) of 0.91 (median modelling efficiency (ME): 0.75). Although the crop biomass, habitus and soil moisture variables were all within an acceptable range, the model often underperformed for variables related to nitrogen. Uncalibrated MONICA simulations yielded a median nMAE of 0.27 across all observed target variables (n = 85) and a median d of 0.76 (median ME: 0.30), also showing predominantly acceptable results for the crop biomass, habitus and soil moisture variables. Based on the convincing performance of the model under uncalibrated conditions, MONICA can be regarded as a suitable simulation model for use in regional applications. Furthermore, its ability to reproduce the observed crop growth results in free-air carbon enrichment experiments makes it suited to predict agro-ecosystem behaviour under expected future climate conditions.     

Coupling a land use model and an ecosystem model for a crop-pasture zone

This paper describes the development of a land use model coupling ecosystem processes. For a given land use pattern in a region, a built-in regional ecosystem model (TESim) simulates leaf physiology of plants, carbon and nitrogen dynamics, and hydrological processes including runoff generation and run-on re-absorption, as well as runoff-induced soil erosion and carbon and nitrogen loss from ecosystems. The simulation results for a certain period from 1976 to 1999 were then used to support land use decisions and to assess the impacts of land use changes on environment. In the coupling model, the land use type for a land unit was determined by optimization of a weighted suitability derived from expert knowledge about the ecosystem state and site conditions. The model was applied to the temperate crop-pasture band in northern China (CCPB) to analyze the interactions between land use and major ecosystem processes and functions and to indicate the added value of the feedbacks by comparing simulations with and without the coupling and feedbacks between land use module and ecosystem processes. The results indicated that the current land use in CCPB is neither economical nor ecologically judicious. The scenario with feedbacks increased NPP by 46.78 g C m⁻² a⁻¹, or 32.23% of the scenario without feedbacks, also decreased soil erosion by 0.65 kg m⁻² a⁻¹, or 23.13%. Without altering the regional land use structure (proportions of each land use type). The system developed in this study potentially benefits both land managers and researchers.     

Utilizing satellite imagery and GLOBE student data to model soil dynamics

General Purpose Atmosphere Plant Soil Simulator (GAPS), a menu-driven soil-vegetation-atmosphere transfer (SVAT) model, was used to simulate soil water dynamics from 1998 through 2001 for Greenville, PA, USA. GLOBE student data collected by students from Reynolds Junior and Senior High School, coupled with normalized difference vegetation index (NDVI) data derived from SPOT4 vegetation imagery, were used to parameterize and validate the model. Data from the National Weather Service Cooperative (NWSC) was used to evaluate the GLOBE dataset. Overall, there was a high index of agreement (d > 0.80) between field measurements and simulated soil water values from both datasets (GLOBE and NWSC). Simulations using the GLOBE climate data outperformed the NWSC data for the 1999, 2000, and 2001 growing seasons. In addition, the GLOBE simulations showed that NDVI could be utilized to predict transpiration periods (QI, QII, and QIII) for northern latitudes >35° with a distinct winter period. In phenological terms, QI reflects the onset of the growing season when vegetation is greening up (NDVI < 0.60) and transpiration is beginning (<2 mm/day) and QII reflects the end of the growing seasons when vegetation is greening down and transpiration is decreasing. QIII reflects the height of the growing season when transpiration rates average between 2 and 5 mm per day and NDVI is at its maximum (>0.60). Results of this study demonstrate that GLOBE student data, coupled with remotely sensed data, can provide an important source of input and validation information for capacitance SVAT models such as GAPS.     

Vegetation competition model for water and light limitation. I: Model description, one-dimensional competition and the influence of groundwater

Vegetation growth models often concentrate on the interaction of vegetation with soil moisture but usually omit the influence of groundwater. However the proximity of groundwater can have a profound effect on vegetation growth, because it strongly influences the spatial and temporal distribution of soil moisture and therefore water and oxygen stress of vegetation. In two papers we describe the behavior of a coupled vegetation-groundwater-soil water model including the competition for water and light. In this first paper we describe the vegetation model, compare the model to measured flux data and show the influence of water and light competition in one dimension. In the second paper we focus on the influence of lateral groundwater flow and spatial patterns along a hillslope. The vegetation model is based on a biophysical representation of the soil-plant-atmosphere continuum. Transpiration and stomatal conductance depend both on atmospheric forcing and soil moisture content. Carbon assimilation depends on environmental conditions, stomatal conductance and biochemical processes. Light competition is driven by tree height and water competition is driven by root water uptake and its water and oxygen stress reaction. The modeled and measured H₂O and CO₂ fluxes compare well to observations on both a diurnal and a yearly timescale. Using an upscaling procedure long simulation runs were performed. These show the importance of light competition in temperate forests: once a tree is established under slightly unfavorable soil moisture conditions it can not be outcompeted by smaller trees with better soil moisture uptake capabilities, both in dry as in wet conditions. Performing the long simulation runs with a background mortality rate reproduces realistic densities of wet and dry adapted tree species along a wet to dry gradient. These simulations show that the influence of groundwater is apparent for a large range of groundwater depths, by both capillary rise and water logging. They also show that species composition and biomass have a larger influence on the water balance in eco-hydrological systems than soil and groundwater alone.     

SPACSYS: Integration of a 3D root architecture component to carbon, nitrogen and water cycling—Model description

It is an ongoing challenge to develop and demonstrate management practices that increase the sustainability of agricultural systems. Soil carbon and nitrogen dynamics directly affect soil quality, crop productivity and environmental impacts. Root systems are central to the acquisition of water and nutrients by plants, but are also a major pathway for the inputs of carbon and nutrients to soil. The complexity of both biotic and abiotic interactions, combined with stochastic changes in root architecture, makes it difficult to understand below-ground dynamics on the basis of experimentation alone. The integration of dynamic models of above-ground growth, three-dimensional root system demography, and interactions between plants and the environment, into one single model is a major challenge because of the complexity of the systems.In order to understand the interaction between a plant and the environment, it is advantageous to develop a model framework to integrate submodels that simulate various plant and environmental components. The objective of this paper is to outline a mechanistic and process-based model, which is capable of simulating interactions among environmental conditions around plants, plant growth and development, nitrogen and carbon cycles, with a three-dimensional root system submodel as an interface.The model presented in this paper is a mixed dimensional, multi-layer, field scale, weather-driven and daily time-step dynamic simulation model. The current version includes a plant growth and development component, a nitrogen cycling component, a carbon cycling component, plus a soil water component that includes representation of water flow to field drains as well as downwards through the soil layers, together with a heat transfer component. The components themselves and linkage among components are designed using object-oriented techniques, which makes the model robust, understandable and reusable. The components are implemented in the C++ programming language, and inputs and outputs of all components are organised as a database in either Microsoft^® SQL Server 2000, Access 2000 or MySQL5.0. Root architecture is visualised by using the OpenGL graphics system. Preliminary validation with two separate experimental datasets shows that the model can reasonably simulate root systems, nitrogen cycling, water movement and plant growth.     

Water budget and the consequent duration of canopy carbon gain in a teak plantation in a dry tropical region: Analysis using a soil–plant–air continuum multilayer model

A soil–plant–air continuum multilayer model was used to numerically simulate canopy net assimilation (A_n), evapotranspiration (ET), and soil moisture in a deciduous teak plantation in a dry tropical climate of northern Thailand to examine the influence of soil drought on A_n. The timings of leaf flush and the end of the canopy duration period (CDP) were also investigated from the perspective of the temporal positive carbon gain. Two numerical experiments with different seasonal patterns of leaf area index (LAI) were carried out using above-canopy hydrometeorological data as input data. The first experiment involved seasonally varying LAI estimated based on time-series of radiative transmittance through the canopy, and the second experiment applied an annually constant LAI. The first simulation captured the measured seasonal changes in soil surface moisture; the simulated transpiration agreed with seasonal changes in heat pulse velocity, corresponding to the water use of individual trees, and the simulated A_n became slightly negative. However, in the second simulation, A_n became negative in the dry season because the decline in stomatal conductance due to severe soil drought limited the assimilation, and the simultaneous increase in leaf temperature increased dark respiration. Thus, these experiments revealed that the leaflessness in the dry season is reasonable for carbon gain and emphasized the unfavorable soil water status for carbon gain in the dry season. Examining the duration of positive A_n (DPA) in the second simulation showed that the start of the longest DPA (LDPA) in a year approached the timing of leaf flush in the teak plantation after the spring equinox. On the other hand, the end appeared earlier than that of all CDPs. This result is consistent with the sap flow stopping earlier than the complete leaf fall, implying that the carbon assimilation period ends before the completion of defoliation. The model sensitivity analysis in the second simulation suggests that a smaller LAI and slower maximum rate of carboxylation likely extend the LDPA because soil water from the surface to rooting depth is maintained longer at levels adequate for carbon gain by decreased canopy transpiration. The experiments also suggest that lower soil hydraulic conductivity and deeper rooting depth can postpone the end of the LDPA by increasing soil water retention and the soil water capacity, respectively.     

An architectural model of spring wheat: Evaluation of the effects of population density and shading on model parameterization and performance

ADELwheat is an architectural model that describes development of wheat in 3D. This paper analyzes the robustness of the parameterization of ADELwheat for spring wheat cultivars in relation to plant population density and shading. The model was evaluated using data from two spring wheat experiments with three plant population densities and two light regimes. Model validation was done at two levels of aggregation: (a) by comparing parameterization functions used as well as parameter values to the data (leaf and tiller appearance, leaf number, blade dimensions, sheath length, internode length) and (b) by comparing ground cover (GC) and leaf area index (LAI) of simulated virtual wheat plots with GC and LAI calculated from data. A sensitivity analysis was performed by modulating parameters defining leaf blade dimensions and leaf or tiller appearance rate.In contrast to population density, shading generally increased phyllochron and delayed tiller appearance. Both at the level of the organ and at the level of the canopy the model performed satisfactorily. Parameterization functions in the model that had been established previously applied to independent data for different conditions; GC and LAI were simulated adequately at three population densities. Sensitivity analysis revealed that calibration of phyllochron and blade area needs to be accurate to prevent disproportional deviations in output.The robustness of the model parameterization and the simulation performance confirmed that the model is a complete architectural model for aboveground development of spring wheat. It can be used in studies that require simulation of spring wheat structure, such as studies on plant–insect interaction, remote sensing, and light interception.     

Dynamics of soil organic matter in primary and secondary forest succession on sandy soils in The Netherlands: An application of the ROMUL model

We applied the simulation model ROMUL of soil organic matter dynamics in order to analyse and predict forest soil organic matter (SOM) changes following stand growth and also to identify gaps of data and modelling problems. SOM build-up was analysed (a) from bare sand to forest soil during a primary succession in Scots pine forest and (b) on mature forest soil under Douglas fir plantations as an example of secondary succession in The Netherlands. As some of the experimental data were unreliable we compiled a set of various scenarios with different soil moisture regime, initial SOM pools and amount and quality of above and below ground litter input. This allowed us to find the scenarios that reflect the SOM dynamics more realistically. In the Scots pine forest, total litter input was estimated as 0.50 kg m⁻² year⁻¹. Two scenarios were defined for the test runs: (a) forest floor moisture regimes—‘dry, mesic and hydric’ and (b) augmenting a root litter pool with three ratios of needles and branches to roots: 1:1, 1:1.5 and 1:2.0. The scenario finally compiled had the following characteristics: (a) climate for dry site with summer drought and high winter moisture of forest floor; (b) a litter input of 0.25 kg m⁻² year⁻¹ above ground and 0.50 kg m⁻² year⁻¹ below ground; (c) a low nitrogen and ash content in all litter fall fractions. The test runs for the estimation of the initial SOM pools and the amount and proportion of above and below ground litter fall were also performed in the Douglas fir plantation. The inputs of above ground litter tested in various combinations were 0.30 and 0.60 kg m⁻² year⁻¹, and below ground litter 0.30, 0.60 and 0.90 kg m⁻² year⁻¹. The scenario that fitted the experimental data had an SOM pool of 20–25 kg m⁻², an aboveground litter input of 0.6 kg m⁻² year⁻¹and a below ground litter input of 0.9 kg m⁻² year⁻¹. The long-term simulation corresponded well with the observed patterns of soil organic matter accumulation associated with the forest soil development in primary and secondary succession. During primary succession in Scots pine forest on dry sand there is a consistent accumulation of a raw humus forest floor. The soil dynamics in the Douglas fir plantation also coincide with the observed patterns of SOM changes during the secondary succession, with SOM decreasing significantly under young forest, and SOM being restored in the older stands.     

A neural network experiment on the simulation of daily nitrate-nitrogen and suspended sediment fluxes from a small agricultural catchment

In this paper, we report an application of neural networks to simulate daily nitrate-nitrogen and suspended sediment fluxes from a small 7.1 km² agricultural catchment (Melarchez), 70 km east of Paris, France. Nitrate-nitrogen and sediment losses are only a few possible consequences of soil erosion and biochemical applications associated to human activities such as intensive agriculture. Stacked multilayer perceptrons models (MLPs) like the ones explored here are based on commonly available inputs and yet are reasonably accurate considering their simplicity and ease of implementation. Note that the simulation does not resort on water quality flux observations at previous time steps as model inputs, which would be appropriate, for example, to predict the water chemistry of a drinking water plant a few time steps ahead. The water quality fluxes are strictly mapped to historical mean flux values and to hydro-climatic variables such as stream flow, rainfall, and soil moisture index (12 model input candidates in total), allowing its usage even when no flux observations are available. Self-organizing feature maps based on the network structure established by Kohonen were employed first to produce the training and the testing data sets, with the intent to produce statistically close subsets so that any difference in model performance between validation and testing has to be associated to the model and not to the data subsets. The stacked MLPs reached different levels of performance simulating the nitrate-nitrogen flux and the suspended sediment flux. In the first instance, 2-input stacked MLP nitrate-nitrogen simulations, based on the same-day stream flow and on the 80-cm soil moisture index, have a performance of almost 90% according to the efficiency index. On the other hand, the performance of 3-input stacked MLPs (same-day stream flow, same-day historical flux, and same-day stream flow increment) reached a little more than 75% according to the same criterion. The results presented here are deemed already promising enough, and should encourage water resources managers to implement simple models whenever appropriate.     

Contaminant transport in an unsaturated soil: laboratory tests and numerical simulation model as procedure for parameters evaluation

In this paper, lab tests coupled to a semi-pilot test section are used to derive data for the calibration of a numerical model. The paper is aimed at proposing a set of experiments, which can be used to calibrate a numerical model before using it on defined soils. The complexity of the phenomenon of transport of reactive pollutants in soil has to be faced in the most complete way. The different behaviour of soil after wet/dry cycles with respect to the fluidodynamic characteristics and the importance to consider the local biomass behaviour in case of organic contaminant has been underlined. An optimal approach has to take into account all the different components and here a simple series of experimental procedure is presented. The sensitivity analysis of the numerical model has shown that its results are not so much dependent on the classical numerical aspects (time or space increments) but mainly on a set of parameters related to soil structure which must then be derived through a good calibration.     

Sensitivity of species-distribution models to error,bias, and model design: An application to resource selection functions for woodland caribou

Models that predict distribution are now widely used to understand the patterns and processes of plant and animal occurrence as well as to guide conservation and management of rare or threatened species. Application of these methods has led to corresponding studies evaluating the sensitivity of model performance to requisite data and other factors that may lead to imprecise or false inferences. We expand upon these works by providing a relative measure of the sensitivity of model parameters and prediction to common sources of error, bias, and variability. We used a one-at-a-time sample design and GPS location data for woodland caribou (Rangifer tarandus caribou) to assess one common species-distribution model: a resource selection function. Our measures of sensitivity included change in coefficient values, prediction success, and the area of mapped habitats following the systematic introduction of geographic error and bias in occurrence data, thematic misclassification of resource maps, and variation in model design. Results suggested that error, bias and model variation have a large impact on the direct interpretation of coefficients. Prediction success and definition of important habitats were less responsive to the perturbations we introduced to the baseline model. Model coefficients, prediction success, and area of ranked habitats were most sensitive to positional error in species locations followed by sampling bias, misclassification of resources, and variation in model design. We recommend that researchers report, and practitioners consider, levels of error and bias introduced to predictive species-distribution models. Formal sensitivity and uncertainty analyses are the most effective means for evaluating and focusing improvements on input data and considering the range of values possible from imperfect models.     

A remote sensing-based primary production model for grassland biomes

That data from polar orbiting satellites have detected a widespread increase in photosynthetic activity over the last 20 years in the grasslands of the Sahel is justifies investigating its role in the tropical carbon cycle. But this task is undermined because ground data that are generally used to support the use of primary production models elsewhere are lacking. In this paper, we profile a Light Use Efficiency (LUE) model of primary production parameterised with satellite information, and test it for the West African Sahel; solar radiation is absorbed by plants to provide energy for photosynthesis, while moisture shortfalls control the efficiency of light usage. In particular, we show how an economical use of existing, yet meagre data sets can be used to circumvent nominal, yet untenable approaches for achieving this for the region. Specifically, we use a cloudiness layer provided with the NOAA/NASA 8 km Pathfinder Land data archive (PAL) data set to derive solar radiation (and other energy balance terms) required to implement the model (monthly time-step). Of particular note, we index growth efficiency via transpiration by subsuming rangeland-yield formulations into our model. This is important for partially vegetated landscapes where the fate of rainfall is controlled by relative vegetation cover. We accomplish this by using PAL-derived Normalised Difference Vegetation Index (NDVI) to partition the landscape into fractional vegetation cover. A bare soil evaporation model that feeds into bucket model is then applied, thereafter deriving actual transpiration (quasi-daily time-step). We forgo a formal validation of the model due to problems of spatial scale and data limitations. Instead, we generate maps showing model robustness via Monte Carlo simulation. The precision of our Gross Primary Production (GPP) estimates is acceptable, but falls off rapidly for the northern fringes of the Sahel. We also map the locations where errors in the driving variables are mostly responsible for the bulk of uncertainty in predicted GPP, in this case the water stress factor and the NDVI. Comparisons with an independent model of primary production, CENTURY, are relatively poor, yet favourable comparisons are made with previous primary production estimates found for the region in the literature. A spatially exhaustive evaluation of our GPP map is carried out by regressing randomly sampled observations against integrated NDVI, a method traditionally used to quantify absolute amounts of primary production. Our model can be used to quantify stocks and flows of carbon in grasslands over the recent historical period.