Space-time modeling of environmental monitoring data

This study describes and applies statistical methods for space-time modeling of data from environmental monitoring programs, e.g., within areas such as climate change, air pollution and aquatic environment. Such data are often characterized by sparse sampling in both the temporal and spatial dimensions. In order to improve the amount of information on the physical system in question we suggest using statistical modeling methods for monitoring data. Model predictions combined with observations could be analyzed directly to assess the environmental state or as forcing functions for time series models and deterministic, hydrodynamic models. To illustrate the approach we applied the proposed modeling methods to data from the Danish and Swedish marine monitoring programs. Time series with a weekly resolution were predicted from observations of dissolved inorganic nitrogen (DIN) from the Kattegat basin (1993–1997). DIN observations were sparse, irregularly distributed and comprised approximately 10% of the generated time series.     

Spatial prediction of soil contamination based on machine learning: a review

● A review of machine learning (ML) for spatial prediction of soil contamination. ● ML have achieved significant breakthroughs for soil contamination prediction. ● A structured guideline for using ML in soil contamination is proposed. ● The guideline includes variable selection, model evaluation, and interpretation. Soil pollution levels can be quantified via sampling and experimental analysis; however, sampling is performed at discrete points with long distances owing to limited funding and human resources, and is insufficient to characterize the entire study area. Spatial prediction is required to comprehensively investigate potentially contaminated areas. Consequently, machine learning models that can simulate complex nonlinear relationships between a variety of environmental conditions and soil contamination have recently become popular tools for predicting soil pollution. The characteristics, advantages, and applications of machine learning models used to predict soil pollution are reviewed in this study. Satisfactory model performance generally requires the following: 1) selection of the most appropriate model with the required structure; 2) selection of appropriate independent variables related to pollutant sources and pathways to improve model interpretability; 3) improvement of model reliability through comprehensive model evaluation; and 4) integration of geostatistics with the machine learning model. With the enrichment of environmental data and development of algorithms, machine learning will become a powerful tool for predicting the spatial distribution and identifying sources of soil contamination in the future.     

Parameterization of Albedo over Heterogeneous Surfaces in Coupled Land-Atmosphere Schemes for Environmental Modeling. Part I: Theoretical Background

In grid-based environmental models, the underlying surface consists of patches of solid and liquid parts and different plant communities, creating a very heterogeneous picture in the grid cell. In these cases, numerical modelers usually use a simple arithmetic average to determine the grid-cell albedo, a key variable in the parameterization of the land-surface radiative transfer over the grid cell. The object of this paper is to consider the assumptions for aggregating the albedo over a very heterogeneous surface where various surfaces occur at different heights, and, then propose a method for deriving a general expression for it. The suggested expression for the albedo is compared with the conventional approach, for the two-patches grid-cell with a simple geometrical distribution and different heights of its components. A numerical test is performed to compare the two approaches by numerical simulation of the evolution of the surface temperature over the particular grid-cell. Specifically, a one-dimensional land-surface model was applied to an isolated rocky grid-cell with a hole in the center; the model was forced with meteorological observations taken on July 17, 1999 in Philadelphia, PA.     

Scale-dependent relations in land cover biophysical dynamics

The exploration of the relationships between plant biotic dynamics and scale can reveal important information on ecosystem spatial organization by addressing preservation of information integrity in upscaling/downscaling procedures of land-surface parameterization for environmental modeling applications. Scale-dependent relations of vegetation dynamics are investigated in this study by using emergent biophysical characteristics obtained through a predictive multidimensional model of vegetation anomalies derived from remote-sensing observations. In particular, the analysis is focused on the spatial organization of some phenological parameters including deterministic variations (seasonal range, interannual variability, jump discontinuities) and stochastic components (plant memory, spatial correlations). The analysis is performed using MODIS-based Normalized Difference Vegetation Index (NDVI) 16-day composites for the period from March 2000 to December 2006 over Italy at different levels of spatial aggregation (1-8 km). Scale-dependences of the statistical moments of the phenological parameters are quantified through simple power laws for five distinct vegetated land covers. Results suggest that some biophysical characteristics, especially deterministic components, show no preferential spatial scale for important coverage. In particular, broad-leaved forests and natural grasslands are characterized by deterministic and low-distance spatial components well explained by scale relationships, which are modulated by possible spatiotemporal dynamics of climatic drivers. Agricultural lands show high scale-dependent relations on short-term biophysical memory sources and low-distance spatial components of phenology likely related to hierarchical interactions of anthropogenic and ecological processes; whereas mixed patterns of croplands and natural areas generally present no consistent scaling relations.     

Using the ECLPSS software environment to build a spatially explicit component-based model of ozone effects on forest ecosystems

We have developed a modeling framework to support grid-based simulation of ecosystems at multiple spatial scales, the Ecological Component Library for Parallel Spatial Simulation (ECLPSS). ECLPSS helps ecologists to build robust spatially explicit simulations of ecological processes by providing a growing library of reusable interchangeable components and automating many modeling tasks. To build a model, a user selects components from the library, and then writes new components as needed. Some of these components represent specific ecological processes, such as how environmental factors influence the growth of individual trees. Other components provide simulation support such as reading and writing files in various formats to allow inter-operability with other software. The framework manages components and variables, the order of operations, and spatial interactions. The framework provides only simulation support; it does not include ecological functions or assumptions. This separation allows biologists to build models without becoming computer scientists, while computer scientists can improve the framework without becoming ecologists. The framework is designed to operate on multiple platforms and be used across networks via a World Wide Web-based user interface. ECLPSS is designed for use with both single processor computers for small models, and multiple processors in order to simulate large regions with complex interactions among many individuals or ecological compartments. To test Version 1.0 of ECLPSS, we created a model to evaluate the effect of tropospheric ozone on forest ecosystem dynamics. This model is a reduced-form version of two existing models: , which represents an individual tree, and , which represents forest stand growth and succession. This model demonstrates key features of ECLPSS, such as the ability to examine the effects of cell size and model structure on model predictions.     

Integration of a fish bioenergetics model into a spatially explicit water quality model: Application to menhaden in Chesapeake Bay

Although fish are usually thought of as victims of water quality degradation, it has been proposed that some planktivorous species may improve water quality through consumption of algae and sequestering of nutrients via growth. Within most numerical water quality models, the highest trophic level modeled explicitly is zooplankton, prohibiting an investigation of the effect a fish species may be having on its environment. Conversely, numerical models of fish consumption do not typically include feedback mechanisms to capture the effects of fish on primary production and nutrient recycling. In the present study, a fish bioenergetics model is incorporated into CE-QUAL-ICM, a spatially explicit eutrophication model. In addition to fish consumption of algae, zooplankton, and detritus, fish biomass accumulation and nutrient recycling to the water column are explicitly accounted for. These developments advance prior modeling efforts of the impact of fish on water quality, many of which are based on integrated estimates over an entire system and which omit the feedback the fish have through nutrient recycling and excretion. To validate the developments, a pilot application was undertaken for Atlantic menhaden (Brevoortia tyrannus) in Chesapeake Bay. The model indicates menhaden may reduce the algal biomass while simultaneously increasing primary productivity.     

An enhanced environmental multimedia modeling system based on fuzzy-set approach: I. Theoretical framework and model development

Multimedia environmental modeling is extremely complex due to the intricacy of the systems with the consideration of many related factors. Traditional environmental multimedia models (EMMs) are usually based on one-dimensional and first-order assumptions, which may cause numerical errors in the simulation results. In this study, a new user-friendly fuzzy-set enhanced environmental multimedia modeling system (FEEMMS) is developed, and includes four key modules: an air dispersion module, a polluting source module, an unsaturated zone module, and a groundwater module. Many improvements over previous EMMs have been achieved through dynamically quantifying the intermedia mass flux; incorporating fuzzy-set approach into environmental multimedia modeling system (EMMS); and designing a user-friendly graphic user interface (GUI). The developed FEEMMS can be a useful tool in estimating the time-varying and spatial-varying chemical concentrations in air, soil, and groundwater; characterizing the potential risk to human health presented by contaminants released from a contaminated site; and quantifying the uncertainties associated with modeling systems and subsequently providing robustness and flexibility for the remediation-related decision making.     

An integrated modeling framework for performing environmental assessments: Application to ecosystem services in the Albemarle-Pamlico basins (NC and VA, USA)

The U.S. Environmental Protection Agency uses environmental models to inform rulemaking and policy decisions at multiple spatial and temporal scales. As decision-making has moved towards integrated thinking and assessment (e.g. media, site, region, services), the increasing complexity and interdisciplinary nature of modern environmental problems has necessitated a new generation of integrated modeling technologies. Environmental modelers are now faced with the challenge of determining how data from manifold sources, types of process-based and empirical models, and hardware/software computing infrastructure can be reliably integrated and applied to protect human health and the environment.In this study, we demonstrate an Integrated Modeling Framework that allows us to predict the state of freshwater ecosystem services within and across the Albemarle-Pamlico Watershed, North Carolina and Virginia (USA). The Framework consists of three facilitating technologies: Data for Environmental Modeling automates the collection and standardization of input data; the Framework for Risk Assessment of Multimedia Environmental Systems manages the flow of information between linked models; and the Supercomputer for Model Uncertainty and Sensitivity Evaluation is a hardware and software parallel-computing interface with pre/post-processing analysis tools, including parameter estimation, uncertainty and sensitivity analysis. In this application, five environmental models are linked within the Framework to provide multimedia simulation capabilities: the Soil Water Assessment Tool predicts watershed runoff; the Watershed Mercury Model simulates mercury runoff and loading to streams; the Water quality Analysis and Simulation Program predicts water quality within the stream channel; the Habitat Suitability Index model predicts physicochemical habitat quality for individual fish species; and the Bioaccumulation and Aquatic System Simulator predicts fish growth and production, as well as exposure and bioaccumulation of toxic substances (e.g., mercury).Using this Framework, we present a baseline assessment of two freshwater ecosystem services-water quality and fisheries resources-in headwater streams throughout the Albemarle-Pamlico. A stratified random sample of 50 headwater streams is used to draw inferences about the target population of headwater streams across the region. Input data is developed for a twenty-year baseline simulation in each sampled stream using current land use and climate conditions. Monte Carlo sampling (n = 100 iterations per stream) is also used to demonstrate some of the Framework's experimental design and data analysis features. To evaluate model performance and accuracy, we compare initial (i.e., uncalibrated) model predictions (water temperature, dissolved oxygen, fish density, and methylmercury concentration within fish tissue) against empirical field data. Finally, we ‘roll-up’ the results from individual streams, to assess freshwater ecosystem services at the regional scale.     

Multiple-pattern parameter identification and uncertainty analysis approach for water quality modeling

This paper presents a multiple-pattern parameter identification and uncertainty analysis approach for robust water quality modeling using a neural network (NN) embedded genetic algorithm (GA). The modeling approach uses an adaptive NN–GA framework to inversely solve the governing equations in a water quality model for multiple parameter patterns, along with an alternating fitness method to maintain solution diversity. The procedure was demonstrated through a coupled 2D hydrodynamic and eutrophication model for Loch Raven Reservoir in Maryland. The inverse problem was formulated as a nonlinear optimization problem minimizing the degree of misfit (DOM) between model results and observed data. A set of NN models was developed to approximate the input-output response relationship of the Loch Raven Reservoir model and was incorporated into a GA framework in an adaptive fashion to search for near-optimal solutions minimizing the DOM. The numerical example showed that the adaptive NN–GA approach is capable of identifying multiple parameter patterns that reproduce the observed data equally well. The resulting parameter patterns were incorporated into the numerical model, and a multiple-pattern robust water quality modeling analysis, along with a compound margin of safety (CMOS) method, was proposed and applied to analyze the parameter pattern uncertainty.     

Integrated model projections of climate change impacts on a North American lake

Climate change is likely to impact terrestrial and aquatic ecosystems via numerous physical and biological mechanisms. This study outlines a framework for projecting potential impacts of climate change on lakes using linked environmental models. Impacts of climate drivers on catchment hydrology and thermal balance in Onondaga Lake (New York State) are simulated using mechanistic models HSPF and UFILS4. Outputs from these models are fed into a lake ecosystem model, developed in AQUATOX. Watershed simulations project increases in the magnitude of peak flows and consequent increases in catchment nutrient export as the magnitude of extreme precipitation events increases. This occurs concurrently with a decrease in annual stream discharge as a result of increased evapotranspiration. Simulated lake water temperatures increase by as much as 5 °C during the 2040-2069 time period, accompanied by a prolonging of the duration of summer stratification. Projected changes include shifts in the timing of nutrient cycling between lake sediments and water column. Plankton taxa projected to thrive under climate change include green algae and Bosmina longirostris. Responses for species at higher trophic levels are mixed. Benthic macroinvertebrates may either prosper (zebra mussels) or decline (chironomids), while fish (e.g., gizzard shad) exhibit high seasonal variability without any clear trend.     

Local spatial modeling of white-tailed deer distribution

Complex spatial heterogeneity of ecological systems is difficult to capture and interpret using global models alone. For this reason, recent attention has been paid to local spatial modeling techniques. We used one local modeling approach, geographically weighted regression (GWR), to investigate the effects of local spatial heterogeneity on multivariate relationships of white-tailed deer distribution using land cover patch metrics and climate factors. The results of these analyses quantify differences in the contributions of model parameters to estimates of deer density over space. A GWR model with local kernel bandwidth was compared to a GWR model with global kernel bandwidth and an ordinary least-squares regression (OLS) model with the same parameters to evaluate their relative abilities in modeling deer distributions. The results indicated that the GWR models predicted deer density better than the traditional ordinary least-squares model and also provided useful information regarding local environmental processes affecting deer distribution. GWR model comparisons showed that the local kernel bandwidth GWR model was more realistic than the global kernel bandwidth GWR model, as the latter exaggerated local spatial variation. The parameter estimates and model statistics (e.g., model R²) of the GWR models were mapped using geographic information systems (GIS) to illustrate local spatial variation in the regression relationship and to identify causes of large-scale model misspecifications and low estimation efficiencies.     

Numerical analysis of non-constant pure rate of time preference: A model of climate policy

When current decisions affect welfare in the far-distant future, as with climate change, the use of a declining pure rate of time preference (PRTP) provides potentially important modeling flexibility. The difficulty of analyzing models with non-constant PRTP limits their application. We describe and provide software (available online) to implement an algorithm to numerically obtain a Markov perfect equilibrium for an optimal control problem with non-constant PRTP. We apply this software to a simplified version of the numerical climate change model used in the Stern Review. For our calibration, the policy recommendations are less sensitive to the PRTP than widely believed.     

Source identification and prediction of nitrogen and phosphorus pollution of Lake Taihu by an ensemble machine learning technique

● A machine learning model was used to identify lake nutrient pollution sources. ● XGBoost model showed the best performance for lake water quality prediction. ● Model feature size was reduced by screening the key features with the MIC method. ● TN and TP concentrations of Lake Taihu are mainly affected by endogenous sources. ● Next-month lake TN and TP concentrations were predicted accurately. Effective control of lake eutrophication necessitates a full understanding of the complicated nitrogen and phosphorus pollution sources, for which mathematical modeling is commonly adopted. In contrast to the conventional knowledge-based models that usually perform poorly due to insufficient knowledge of pollutant geochemical cycling, we employed an ensemble machine learning (ML) model to identify the key nitrogen and phosphorus sources of lakes. Six ML models were developed based on 13 years of historical data of Lake Taihu’s water quality, environmental input, and meteorological conditions, among which the XGBoost model stood out as the best model for total nitrogen (TN) and total phosphorus (TP) prediction. The results suggest that the lake TN is mainly affected by the endogenous load and inflow river water quality, while the lake TP is predominantly from endogenous sources. The prediction of the lake TN and TP concentration changes in response to these key feature variations suggests that endogenous source control is a highly desirable option for lake eutrophication control. Finally, one-month-ahead prediction of lake TN and TP concentrations (R² of 0.85 and 0.95, respectively) was achieved based on this model with sliding time window lengths of 9 and 6 months, respectively. Our work demonstrates the great potential of using ensemble ML models for lake pollution source tracking and prediction, which may provide valuable references for early warning and rational control of lake eutrophication.     

Comparison and ranking of different modelling techniques for prediction of site index in Mediterranean mountain forests

Forestry science has a long tradition of studying the relationship between stand productivity and abiotic and biotic site characteristics, such as climate, topography, soil and vegetation. Many of the early site quality modelling studies related site index to environmental variables using basic statistical methods such as linear regression. Because most ecological variables show a typical non-linear course and a non-constant variance distribution, a large fraction of the variation remained unexplained by these linear models. More recently, the development of more advanced non-parametric and machine learning methods provided opportunities to overcome these limitations. Nevertheless, these methods also have drawbacks. Due to their increasing complexity they are not only more difficult to implement and interpret, but also more vulnerable to overfitting. Especially in a context of regionalisation, this may prove to be problematic. Although many non-parametric and machine learning methods are increasingly used in applications related to forest site quality assessment, their predictive performance has only been assessed for a limited number of methods and ecosystems.In this study, five different modelling techniques are compared and evaluated, i.e. multiple linear regression (MLR), classification and regression trees (CART), boosted regression trees (BRT), generalized additive models (GAM), and artificial neural networks (ANN). Each method is used to model site index of homogeneous stands of three important tree species of the Taurus Mountains (Turkey): Pinus brutia, Pinus nigra and Cedrus libani. Site index is related to soil, vegetation and topographical variables, which are available for 167 sample plots covering all important environmental gradients in the research area. The five techniques are compared in a multi-criteria decision analysis in which different model performance measures, ecological interpretability and user-friendliness are considered as criteria.When combining these criteria, in most cases GAM is found to outperform all other techniques for modelling site index for the three species. BRT is a good alternative in case the ecological interpretability of the technique is of higher importance. When user-friendliness is more important MLR and CART are the preferred alternatives. Despite its good predictive performance, ANN is penalized for its complex, non-transparent models and big training effort.     

Flexible hierarchical mark-recapture modeling for open populations using WinBUGS

Hierarchical mark-recapture models offer three advantages over classical mark-recapture models: (i) they allow expression of complicated models in terms of simple components; (ii) they provide a convenient way of modeling missing data and latent variables in a way that allows expression of relationships involving latent variables in the model; (iii) they provide a convenient way of introducing parsimony into models involving many nuisance parameters. Expressing models using the complete data likelihood we show how many of the standard mark-recapture models for open populations can be readily fitted using the software WinBUGS. We include examples that illustrate fitting the Cormack–Jolly–Seber model, multi-state and multi-event models, models including auxiliary data, and models including density dependence.

Exploring community assembly through an individual-based model for trophic interactions

Traditionally, the dynamics of community assembly has been analyzed by means of deterministic models of differential equations. Despite the theoretical advances provided by such models, they are restricted to questions about community-wide features. The individual-based modeling offers an opportunity to link bionomic features to patterns at the community scale, allowing us to understand how trait-based assembly rules can arise by dynamical processes. The present paper introduces an individual-based model of community assembly, and discusses some of the major advantages and drawbacks of this approach. The model was framed to deal with predation among size-structured populations, incorporating allometric constraints to energetic requirements, movement, life-history features and interaction relationships among individuals. A protocol of assembly procedure is proposed, in which a period of intense species introductions is followed by a period without introductions. The resultant communities did not present any pattern of trait over-dispersion, meaning that the multivariate distances of bionomic features among co-occurring species were neither larger nor more regular than expected in a random collection of species. It suggests a weak influence of interspecific interactions in the model environment and individualistic rules of coexistence, driven mainly by the spatial structure. This highlights that trait over-dispersion and resource partitioning should not be considered a necessary condition for coexistence, even in communities entirely structured by internal processes like predation and competition.     

Database-driven models of the world's Large Marine Ecosystems

We present a new methodology for database-driven ecosystem model generation and apply the methodology to the world's 66 currently defined Large Marine Ecosystems. The method relies on a large number of spatial and temporal databases, including FishBase, SeaLifeBase, as well as several other databases developed notably as part of the Sea Around Us project. The models are formulated using the freely available Ecopath with Ecosim (EwE) modeling approach and software. We tune the models by fitting to available time series data, but recognize that the models represent only a first-generation of database-driven ecosystem models. We use the models to obtain a first estimate of fish biomass in the world's LMEs. The biggest hurdles at present to further model development and validation are insufficient time series trend information, and data on spatial fishing effort.     

Maximum entropy modeling of species geographic distributions

The availability of detailed environmental data, together with inexpensive and powerful computers, has fueled a rapid increase in predictive modeling of species environmental requirements and geographic distributions. For some species, detailed presence/absence occurrence data are available, allowing the use of a variety of standard statistical techniques. However, absence data are not available for most species. In this paper, we introduce the use of the maximum entropy method (Maxent) for modeling species geographic distributions with presence-only data. Maxent is a general-purpose machine learning method with a simple and precise mathematical formulation, and it has a number of aspects that make it well-suited for species distribution modeling. In order to investigate the efficacy of the method, here we perform a continental-scale case study using two Neotropical mammals: a lowland species of sloth, Bradypus variegatus, and a small montane murid rodent, Microryzomys minutus. We compared Maxent predictions with those of a commonly used presence-only modeling method, the Genetic Algorithm for Rule-Set Prediction (GARP). We made predictions on 10 random subsets of the occurrence records for both species, and then used the remaining localities for testing. Both algorithms provided reasonable estimates of the species’ range, far superior to the shaded outline maps available in field guides. All models were significantly better than random in both binomial tests of omission and receiver operating characteristic (ROC) analyses. The area under the ROC curve (AUC) was almost always higher for Maxent, indicating better discrimination of suitable versus unsuitable areas for the species. The Maxent modeling approach can be used in its present form for many applications with presence-only datasets, and merits further research and development.     

A fast antibiotic detection method for simplified pretreatment through spectra-based machine learning

• A spectral machine learning approach is proposed for predicting mixed antibiotic. • Pretreatment is far simpler than traditional detection methods. • Performance of the model is compared in different influencing factors. • Spectral machine learning is promising in the detection of complex substances. Antibiotics are widely used in medicine and animal husbandry. However, due to the resistance of antibiotics to degradation, large amounts of antibiotics enter the environment, posing a potential risk to the ecosystem and public health. Therefore, the detection of antibiotics in the environment is necessary. Nevertheless, conventional detection methods usually involve complex pretreatment techniques and expensive instrumentation, which impose considerable time and economic costs. In this paper, we proposed a method for the fast detection of mixed antibiotics based on simplified pretreatment using spectral machine learning. With the help of a modified spectrometer, a large number of characteristic images were generated to map antibiotic information. The relationship between characteristic images and antibiotic concentrations was established by machine learning model. The coefficient of determination and root mean squared error were used to evaluate the prediction performance of the machine learning model. The results show that a well-trained machine learning model can accurately predict multiple antibiotic concentrations simultaneously with almost no pretreatment. The results from this study have some referential value for promoting the development of environmental detection technologies and digital environmental management strategies.