Sampling variability and estimates of density dependence: a composite-likelihood approach

It is well known that sampling variability, if not properly taken into account, affects various ecologically important analyses. Statistical inference for stochastic population dynamics models is difficult when, in addition to the process error, there is also sampling error. The standard maximum-likelihood approach suffers from large computational burden. In this paper, I discuss an application of the composite-likelihood method for estimation of the parameters of the Gompertz model in the presence of sampling variability. The main advantage of the method of composite likelihood is that it reduces the computational burden substantially with little loss of statistical efficiency. Missing observations are a common problem with many ecological time series. The method of composite likelihood can accommodate missing observations in a straightforward fashion. Environmental conditions also affect the parameters of stochastic population dynamics models. This method is shown to handle such nonstationary population dynamics processes as well. Many ecological time series are short, and statistical inferences based on such short time series tend to be less precise. However, spatial replications of short time series provide an opportunity to increase the effective sample size. Application of likelihood-based methods for spatial time-series data for population dynamics models is computationally prohibitive. The method of composite likelihood is shown to have significantly less computational burden, making it possible to analyze large spatial time-series data. After discussing the methodology in general terms, I illustrate its use by analyzing a time series of counts of American Redstart (Setophaga ruticilla) from the Breeding Bird Survey data, San Joaquin kit fox (Vulpes macrotis mutica) population abundance data, and spatial time series of Bull trout (Salvelinus confluentus) redds count data.     

An application of Bayesian variable selection to spatial concurrent linear models

Spatial concurrent linear models, in which the model coefficients are spatial processes varying at a local level, are flexible and useful tools for analyzing spatial data. One approach places stationary Gaussian process priors on the spatial processes, but in applications the data may display strong nonstationary patterns. In this article, we propose a Bayesian variable selection approach based on wavelet tools to address this problem. The proposed approach does not involve any stationarity assumptions on the priors, and instead we impose a mixture prior directly on each wavelet coefficient. We introduce an option to control the priors such that high resolution coefficients are more likely to be zero. Computationally efficient MCMC procedures are provided to address posterior sampling, and uncertainty in the estimation is assessed through posterior means and standard deviations. Examples based on simulated data demonstrate the estimation accuracy and advantages of the proposed method. We also illustrate the performance of the proposed method for real data obtained through remote sensing.     

On spline-based approaches to spatial linear regression for geostatistical data

For spatial linear regression, the traditional approach to capture spatial dependence is to use a parametric linear mixed-effects model. Spline surfaces can be used as an alternative to capture spatial variability, giving rise to a semiparametric method that does not require the specification of a parametric covariance structure. The spline component in such a semiparametric method, however, impacts the estimation of the regression coefficients. In this paper, we investigate such an impact in spatial linear regression with spline-based spatial effects. Statistical properties of the regression coefficient estimators are established under the model assumptions of the traditional spatial linear regression. Further, we examine the empirical properties of the regression coefficient estimators under spatial confounding via a simulation study. A data example in precision agriculture research regarding soybean yield in relation to field conditions is presented for illustration.     

Error propagation in spatial modeling of public health data: a simulation approach using pediatric blood lead level data for Syracuse,New York

Lead poisoning produces serious health problems, which are worse when a victim is younger. The US government and society have tried to prevent lead poisoning, especially since the 1970s; however, lead exposure remains prevalent. Lead poisoning analyses frequently use georeferenced blood lead level data. Like other types of data, these spatial data may contain uncertainties, such as location and attribute measurement errors, which can propagate to analysis results. For this paper, simulation experiments are employed to investigate how selected uncertainties impact regression analyses of blood lead level data in Syracuse, New York. In these simulations, location error and attribute measurement error, as well as a combination of these two errors, are embedded into the original data, and then these data are aggregated into census block group and census tract polygons. These aggregated data are analyzed with regression techniques, and comparisons are reported between the regression coefficients and their standard errors for the error added simulation results and the original results. To account for spatial autocorrelation, the eigenvector spatial filtering method and spatial autoregressive specifications are utilized with linear and generalized linear models. Our findings confirm that location error has more of an impact on the differences than does attribute measurement error, and show that the combined error leads to the greatest deviations. Location error simulation results show that smaller administrative units experience more of a location error impact, and, interestingly, coefficients and standard errors deviate more from their true values for a variable with a low level of spatial autocorrelation. These results imply that uncertainty, especially location error, has a considerable impact on the reliability of spatial analysis results for public health data, and that the level of spatial autocorrelation in a variable also has an impact on modeling results.     

A multiscale hierarchical Markov transition matrix model for generating and analyzing thematic raster maps

A model is described for generating hierarchically scaled spatial pattern as represented in a thematic raster map. The model involves a series of Markov transition matrices, one for each level in the scaling hierarchy. In full generality, the model allows the transition matrices to be different at each level, potentially making available a large number of parameters for landscape characterization. The model is self-similar when the transition matrices are all equal. A method is presented for fitting the model to data that take the form of a single-resolution thematic raster map. Explicit analytic solutions are obtained for the fitted parameters. The fitting method is based on a relationship between the hierarchical transitions in the model and spatial transitions at varying distance scales in the data map, a categorical analogy of the geostatistical variogram.     

Optimal model averaging by data perturbation for spatial noisy data

Geostatistical models play an important role in spatial data analysis, in which model selection is inevitable. Model selection methods, such as AIC and BIC, are popular for selecting appropriate models. In recent years, some model averaging methods, such as smoothed AIC and smoothed BIC, are also applied to spatial data models. However, the corresponding averaging estimators are outperformed by optimal model averaging estimators (Hansen in Econometrica 75:1175–1189, 2007) for the ordinary linear models. Therefore, this paper focuses on the optimal model averaging method for geostatistical models. We propose a weight choice criterion for the model averaging estimator on the basis of the generalized degrees of freedom and data perturbation technique. We further theoretically prove the resultant estimator is asymptotically optimal in terms of the mean squared error, and numerically demonstrate its satisfactory performance. Finally, the proposed method is applied to a mercury data set.     

Nonparametric spatial covariance functions: Estimation and testing

Spatial autocorrelation techniques are commonly used to describe genetic and ecological patterns. To improve statistical inference about spatial covariance, we propose a continuous nonparametric estimator of the covariance function in place of the spatial correlogram. The spline correlogram is an adaptation of a recent development in spatial statistics and is a generalization of the commonly used correlogram. We propose a bootstrap algorithm to erect a confidence envelope around the entire covariance function. The meaning of this envelope is discussed. Not all functions that can be drawn inside the envelope are candidate covariance functions, as they may not be positive semidefinite. However, covariance functions that do not fit, are not supported by the data. A direct estimate of the L₀ spatial correlation length with associated confidence interval is offered and its interpretation is discussed. The spline correlogram is found to have high precision when applied to synthetic data. For illustration, the method is applied to electrophoretic data of an alpine grass (Poa alpina).     

Network-based exploration and visualisation of ecological data

Networks – structured graphs consisting of sets of nodes connected by edges – provide a rich framework for data visualisation and exploratory analyses. Although rarely used for the visualisation of ecological data, networks are well suited to this purpose, including data that one might not normally think of as a network. We present a simple method for transforming a data matrix into network format, and show how this can be used as the basis for interactive exploratory analyses of ecological data.The method is demonstrated using a database of marine zooplankton samples acquired in the Southern Ocean. The network analyses revealed zooplankton community structures that are in good agreement with previously published results. Variations in community structure were observed to be related to the temporal and spatial pattern of sampling, as well as to physical environmental factors such as sea ice cover. The analyses also revealed a number of errors in the data, including taxon identification errors and instrument failures.The method allows the analyst to generate networks from different combinations of variables in the data set, and to examine the effects of varying parameters such as the scales of spatial, temporal, and taxonomic aggregation. This flexibility allows the analyst to rapidly gain a number of perspectives on the data and provides a powerful mechanism for exploration.     

A two-step local smoothing approach for exploring spatio-temporal patterns with application to the analysis of precipitation in the mainland of China during 1986–2005

There has been a growing interest on using local modelling techniques for the analysis of spatio-temporal data because of their powerfulness in extracting the underlying local patterns in the data. In this study, we propose a two-step local smoothing approach to explore spatial patterns and temporal trends of spatio-temporal data via combining the geographically weighted regression and the local polynomial smoothing procedure. The proposed method incorporates both spatial and temporal information into the calibration process and makes it easier to implement visualization of the results. A simulation experiment is conducted to assess the performance of the proposed method and the results show that the method works satisfactorily. A real-world spatio-temporal data set is analyzed to demonstrate the practical usefulness of the method.     

Bayesian entropy for spatial sampling design of environmental data

We develop a spatial statistical methodology to design national air pollution monitoring networks with good predictive capabilities while minimizing the cost of monitoring. The underlying complexity of atmospheric processes and the urgent need to give credible assessments of environmental risk create problems requiring new statistical methodologies to meet these challenges. In this work, we present a new method of ranking various subnetworks taking both the environmental cost and the statistical information into account. A Bayesian algorithm is introduced to obtain an optimal subnetwork using an entropy framework. The final network and accuracy of the spatial predictions is heavily dependent on the underlying model of spatial correlation. Usually the simplifying assumption of stationarity, in the sense that the spatial dependency structure does not change location, is made for spatial prediction. However, it is not uncommon to find spatial data that show strong signs of nonstationary behavior. We build upon an existing approach that creates a nonstationary covariance by a mixture of a family of stationary processes, and we propose a Bayesian method of estimating the associated parameters using the technique of Reversible Jump Markov Chain Monte Carlo. We apply these methods for spatial prediction and network design to ambient ozone data from a monitoring network in the eastern US.     

Influence of Representation Targets on the Total Area of Conservation-Area Networks

Abstract:  Systematic conservation planning typically requires specification of quantitative representation targets for biodiversity surrogates such as species, vegetation types, and environmental parameters. Targets are usually specified either as the minimum total area in a conservation-area network in which a surrogate must be present or as the proportion of a surrogate's existing spatial distribution required to be in the network. Because the biological basis for setting targets is often unclear, a better understanding of how targets affect selection of conservation areas is needed. We studied how the total area of conservation-area networks depends on percentage targets ranging from 5% to 95%. We analyzed 12 data sets of different surrogate distributions from 5 regions: Korea, Mexico, Québec, Queensland, and West Virginia. To assess the effect of spatial resolution on the target-area relationship, we also analyzed each data set at 7 spatial resolutions ranging from 0.01°× 0.01° to 0.10°× 0.10°. Most of the data sets showed a linear relationship between representation targets and total area of conservation-area networks that was invariant across changes in spatial resolution. The slope of this relationship indicated how total area increased with target level, and our results suggest that greater surrogate representation requires significantly more area. One data set exhibited a highly nonlinear relationship. The results for this data set suggest a new method for setting targets on the basis of the functional form of target-area relationships. In particular, the method shows how the target-area relationship can provide a rationale for setting targets solely on the basis of distributional information about surrogates.     

A fair comparison between the spatial scan and the Besag–Newell Disease clustering tests

Routine surveillance of a large geographic region for clusters of adverse health events, particularly cancers, often involves small area health data, possibly controlling for exposure information. Many different methods have been proposed to test for the presence of geographical clusters. Two of the most popular methods are the spatial scan method proposed by Kulldorff and that using a fixed number of cases within scanning circles proposed by Besag and Newell. Although the second test is very popular, it has some difficulties. While the scan test controls for the multiple testing problem, the Besag and Newell test does not. Additionally, the latter method requires the setting of several tuning parameters whose values affect the test performance and are subjectively chosen by the user. This creates a difficulty to make a fair comparison between the two methods and it explains why there have been few formal studies evaluating their relative performances. In this paper, we modify the Besag and Newell test allowing for the control of the error type I probability and compare its power with respect to that of the spatial scan test. We used data sets from a publicly available simulated benchmark. We found that the two methods have similar results, except for clusters located in sparsely populated regions, where the spatial scan method presented a better performance.     

Data assimilation in a simple marine ecosystem model based on spatial biological parameterizations

Assuming that a set of constant parameters fits for marine ecosystem modeling and parameter estimation studies on large space scales is questionable since ecosystem types spanning long distances are quite different. In this study, SeaWiFS chlorophyll-a data are assimilated into a simple NPZD model by the adjoint method in a climatological physical environment provided by FOAM. To improve the assimilation results, different spatial parameterization schemes are utilized. The results show that the values of the selected sensitive parameters are spatially variable and the application of spatial parameterizations can improve the assimilation results significantly.     

An adjoint data assimilation approach for estimating parameters in a three-dimensional ecosystem model

In this paper an ecosystem model, including phytoplankton, zooplankton, nitrate, ammonium, phosphate and detritus, is described. The model is driven by physical fields derived from a three-dimensional physical transport model. Simulation includes nitrate input from a river. Simulated results are then sampled and the sampled data are used in sequential numerical experiments to assess the ability of using an adjoint data assimilation approach for estimating the poorly known parameters of the ecosystem model, such as growth and death rate, half-saturation constant of nutrients, etc. Data with different spatial and temporal resolution over 1 week are assimilated into the ecosystem model. Assimilation of data at 30 grid stations with a sampling interval of 6 h is proved to be adequate for recovering all the parameters of the ecosystem model. Both the spatial and temporal resolution of the data are mutually complementary in the assimilative model. Thus, improvement of either of them can result in improvement of model parameter recoveries. The assimilation of phytoplankton data is essential to recover the model parameters. Phytoplankton is the core of the food web and without the information on phytoplankton, the structure of the ecosystem cannot be constructed correctly. The adjoint method can work well with the noisy data. In the twin experiments with noisy data, the parameters can be recovered but the error is increased. The results of the model and parameter recovery are sensitive to the initial conditions of state variables, so the determination of the initial condition is as important as that of the model parameter. The spatial and temporal resolution and the data type of the observations in Analysis and Modelling Research of the Ecosystem in the Bohai Sea (AMREB) are suitable for the recovery of the model parameters used in this study.     

A statistical evaluation of non-ergodic variogram estimators

Geostatistics is a set of statistical techniques that is increasingly used to characterize spatial dependence in spatially referenced ecological data. A common feature of geostatistics is predicting values at unsampled locations from nearby samples using the kriging algorithm. Modeling spatial dependence in sampled data is necessary before kriging and is usually accomplished with the variogram and its traditional estimator. Other types of estimators, known as non-ergodic estimators, have been used in ecological applications. Non-ergodic estimators were originally suggested as a method of choice when sampled data are preferentially located and exhibit a skewed frequency distribution. Preferentially located samples can occur, for example, when areas with high values are sampled more intensely than other areas. In earlier studies the visual appearance of variograms from traditional and non-ergodic estimators were compared. Here we evaluate the estimators' relative performance in prediction. We also show algebraically that a non-ergodic version of the variogram is equivalent to the traditional variogram estimator. Simulations, designed to investigate the effects of data skewness and preferential sampling on variogram estimation and kriging, showed the traditional variogram estimator outperforms the non-ergodic estimators under these conditions. We also analyzed data on carabid beetle abundance, which exhibited large-scale spatial variability (trend) and a skewed frequency distribution. Detrending data followed by robust estimation of the residual variogram is demonstrated to be a successful alternative to the non-ergodic approach.     

Variation partitioning involving orthogonal spatial eigenfunction submodels

When partitioning the variation of univariate or multivariate ecological data with respect to several submodels of spatial eigenfunctions (e.g., Moran's eigenvector maps, MEM) acting as explanatory data, a problem occurs: although the submodels are constructed to be orthogonal to one another, the partitioning based on adjusted R2 statistics produces nonzero values in the intersections between spatial submodels. This phenomenon is described and two solutions are proposed. The first solution is to apportion the intersection fractions proportionally to the variation explained by each submodel. The second solution consists in creating a hierarchy among the spatial submodels, in accordance with hierarchy theory. These solutions lead to new partitioning equations that are described in the Appendix. R functions are provided to carry out partitioning with respect to environmental variables and spatial eigenfunction submodels. This development is important for the correct interpretation of spatial modeling results implying explanatory environmental data as well as submodels of spatial eigenfunctions involving two or more spatial scales.     

Spatial methods for plot-based sampling of wildlife populations

Classical sampling methods can be used to estimate the mean of a finite or infinite population. Block kriging also estimates the mean, but of an infinite population in a continuous spatial domain. In this paper, I consider a finite population version of block kriging (FPBK) for plot-based sampling. The data are assumed to come from a spatial stochastic process. Minimizing mean-squared-prediction errors yields best linear unbiased predictions that are a finite population version of block kriging. FPBK has versions comparable to simple random sampling and stratified sampling, and includes the general linear model. This method has been tested for several years for moose surveys in Alaska, and an example is given where results are compared to stratified random sampling. In general, assuming a spatial model gives three main advantages over classical sampling: (1) FPBK is usually more precise than simple or stratified random sampling, (2) FPBK allows small area estimation, and (3) FPBK allows nonrandom sampling designs.     

Spatial distribution of ecological security status assessment of West-Liaohe River based on geographic information system

Eco-security assessment is a hot research area in resource and environmental science, which involves data with much spatial, non-linear, and random features. Geographic information system (GIS), as a useful tool to analyze and manage spatial information, has a superior advantage in this field. A case study in the western part of the Liaohe River featuring a method of eco-security spatial differences (ESSD) based on GIS is developed in this paper. The method includes four steps: 1) developing the pressure-state-response (P-S-R) framework with site data; 2) digitizing West-Liaohe River and setting its GRID database of ecosecurity assessment indicators; 3) figuring out the relative membership degree (RMD) of eco-security indicators by using the analytical hierarchy process with the weight of indicator; 4) classifying the security zone and mapping the assessment result of eco-security status in grid by GIS method of assigning and clustering. The visual spatial differences of eco-security based on GIS enables decision makers to know the status of eco-security better in making policies for achieving sustainability.     

Spatial variogram estimation from temporally aggregated seabird count data

Seabird abundance is an important indicator for assessing impact of human activities on the marine environment. However, data collection at sea is time consuming and surveys are carried out over several consecutive days for efficiency reasons. This study investigates the validity of aggregating those data over time to estimate a spatial variogram that is representative for spatial correlation in species abundance. For this purpose we simulate four-day surveys of seabird count data that contain spatial and temporal correlation arising from temporal changes in the spatial pattern of environmental conditions. Estimates of the aggregated spatial variogram are compared to a variogram that would arise when data were collected over a single day. The study reveals that, under changing environmental conditions over surveys days, aggregating data over a four-day survey increases both the non-spatial variation in the data and the scale of spatial correlation in seabird data. Next, the effect of using an aggregated variogram on the statistical power to test the significance of an impact is investigated. The impact concerns a case of establishing an offshore wind farm resulting in seabird displacement. The study shows that both overestimation and underestimation of statistical power occurs, with power estimates differing up to a factor of two. We conclude that the spatial variation in seabird abundance can be misrepresented by using temporally aggregated data. In impact studies, such misrepresentation can lead to erroneous assessments of the ability to detect impact.     

Modelling the distribution of plant species using the autologistic regression model

For modeling the distribution of plant species in terms of climate covariates, we consider an autologistic regression model for spatial binary data on a regularly spaced lattice. This model belongs to the class of autologistic models introduced by Besag (1974). Three estimation methods, the coding method, maximum pseudolikelihood method and Markov chain Monte Carlo method are studied and comparedvia simulation and real data examples. As examples, we use the proposed methodology to model the distributions of two plant species in the state of Florida.