Benchmarking a triplet of official estimates

The publication of official statistics at different levels of aggregation requires a benchmarking step. Difficulties arise when a benchmarking method needs to be applied to a triplet of related estimates, at multiple stages of aggregation. For ratios of totals, external benchmarking constraints for the triplet (numerator, denominator, ratio) are that the weighted sum of denominator/numerator/ratio estimates equals to a constant. The benchmarking weight, applied to the ratio estimates, is a function of the denominator estimates. For example, the United States Department of Agriculture’s National Agricultural Statistics Service’s county-level, end-of-season acreage, production and yield estimates need to aggregate to the corresponding agricultural statistics district-level estimates, which also need to aggregate to the corresponding prepublished state-level values. Moreover, the definition of yield, as the ratio of production to harvested acreage, needs to hold at the county level, the agricultural statistics district level and the state level. We discuss different methods of applying benchmarking constraints to a triplet (numerator, denominator, ratio), at multiple stages of aggregation, where the denominator and the ratio are modeled and the numerator is derived. County-level and agricultural statistics district-level, end-of-season acreage, production and yield estimates are constructed and compared using the different methods. Results are illustrated for 2014 corn and soybean in Indiana, Iowa and Illinois.     

Population growth in snow geese: a modeling approach integrating demographic and survey information

There are few analytic tools available to formally integrate information coming from population surveys and demographic studies. The Kalman filter is a procedure that facilitates such integration. Based on a state-space model, we can obtain a likelihood function for the survey data using a Kalman filter, which we may then combine with a likelihood for the demographic data. In this paper, we used this combined approach to analyze the population dynamics of a hunted species, the Greater Snow Goose (Chen caerulescens atlantica), and to examine the extent to which it can improve previous demographic population models. The state equation of the state-space model was a matrix population model with fecundity and regression parameters relating adult survival and harvest rate estimated in a previous capture-recapture study. The observation equation combined the output from this model with estimates from an annual spring photographic survey of the population. The maximum likelihood estimates of the regression parameters from the combined analysis differed little from the values of the original capture-recapture analysis, though their precision improved. The model output was found to be insensitive to a wide range of coefficient of variation (CV) in fecundity parameters. We found a close match between the surveyed and smoothed population size estimates generated by the Kalman filter over an 18-year period, and the estimated CV of the survey (0.078-0.150) was quite compatible with its assumed value (approximately 0.10). When we used the updated parameter values to predict future population size, the model underestimated the surveyed population size by 18% over a three-year period. However, this could be explained by a concurrent change in the survey method. We conclude that the Kalman filter is a promising approach to forecast population change because it incorporates survey information in a formal way compared with ad hoc approaches that either neglect this information or require some parameter or model tuning.     

A grassland ecosystem model of the Xilingol steppe, Inner Mongolia, China

The natural grassland ecosystem of the Xilingol steppe has traditionally been the source of the most productive and highest quality agriculture in northern China. Unfortunately, the area is now experiencing degradation due to resource overuse. In an attempt to forecast grassland production and to sustain the ecosystem, we built a time-dependent simulation model of the ecosystem based on long-range weather forecasts (several weeks to several months). The model incorporated five state variables including above- and belowground biomass, the amount of standing dead plant material, livestock (sheep) weight, and the amount of excrement per unit ground area. Within the model, solar light energy is fixed by grassland vegetation and flows through the other variables via a variety of organism-environment interactions. The model was written using a set of simultaneous differential equations and was numerically analyzed. The values of the time-dependent parameters controlling energy flow were determined based on data accumulated in experiments and field surveys executed at a grassland experimental station located in Xilingol, as well as by reference to related literature. We used daily meteorological data including air temperature and rainfall recorded at the Xilinhot Meteorological Observatory. Simulated results for several stocking densities coincided well with the data of aboveground plant biomass observed at the experimental station in 1990, 1993, and 1997. We obtained reasonable simulation results for five stocking densities, three air temperature patterns, and five rainfall patterns. When a month-long drought, which sometimes occurs in this area, was forecast by a local weather station, a decrease in grassland production was forecast by the model. Such forecasts will assist in the management of livestock, forage preservation, and grassland conservation.     

Hierarchical Bayes estimation of hunting success rates

Although most post-season harvest surveys are conducted at the state level, the effective management of wildlife populations often requires estimates of hunting success rate, hunting pressure and harvest at the sub-area (such as management unit, regional, or county) level.Sample sizes for some sub-areas are often very small or even zero. Because of small sample sizes, estimates for small sub-areas often yield unacceptably large standard errors. In this article, a hierarchical Bayes model is used to estimate hunting success rates at the sub-area level from post-season harvest surveys. The computation is done by Gibbs sampling and adaptive rejection sampling techniques. The method is illustrated using data from the Missouri Turkey Hunting Survey 1994 Spring Season. The Bayesian estimates are close to the frequency estimates for the sub-areas with large sample sizes and more stable than the frequency estimates for those with small sample sizes. The Bayesian estimates will be more useful to wildlife biologists in estab-lishing hunting regulation on small sub-areas at no additional survey cost.     

Line transect sampling of Karner blue butterflies ( Lycaeides melissa samuelis)

Line transect sampling is an effective survey method for estimating butterfly densities because it provides unbiased estimates of site-density (provided key assumptions are met), and estimates are comparable among sites. For monitoring Karner blue butterflies in Wisconsin, USA, comparable estimates are required because each year a different selection of sites will be monitored. Annual state-wide indices of species abundance can be derived from the site-surveys and compared to previous year's indices to monitor trends. We advocate that line transect sampling is preferable to Pollard-Yates transects as a survey technique for monitoring Karner blue butter- flies. The Pollard-Yates surveys do not adjust for diferences in site detectability. As a consequence, estimates of among-site from Pollard-Yates surveys can be biased. © Rapid Science 1998     

Three reasons to use annual payments in contingent valuation surveys: Convergent validity,discount rates,and mental accounting

We present three arguments for using ongoing annual payments in contingent valuation (CV) surveys that estimate the benefit of a long-lasting environmental improvement. First, by matching the duration of the payments with the duration of the environmental benefits, survey respondents are spared from performing complicated present value calculations. Second, willingness to pay (WTP) estimates from CV surveys that include ongoing annual payments best match WTP estimates obtained using travel cost surveys. Third, respondents are less likely to face binding mental budget constraints with ongoing annual payments than with a larger one-time payment. In addition, respondents’ discount rates may be estimated by collecting non-hypothetical, individual time preference data as part of the valuation survey.     

Effects of sample survey design on the accuracy of classification tree models in species distribution models

We evaluated the effects of probabilistic (hereafter DESIGN) and non-probabilistic (PURPOSIVE) sample surveys on resultant classification tree models for predicting the presence of four lichen species in the Pacific Northwest, USA. Models derived from both survey forms were assessed using an independent data set (EVALUATION). Measures of accuracy as gauged by resubstitution rates were similar for each lichen species irrespective of the underlying sample survey form. Cross-validation estimates of prediction accuracies were lower than resubstitution accuracies for all species and both design types, and in all cases were closer to the true prediction accuracies based on the EVALUATION data set. We argue that greater emphasis should be placed on calculating and reporting cross-validation accuracy rates rather than simple resubstitution accuracy rates. Evaluation of the DESIGN and PURPOSIVE tree models on the EVALUATION data set shows significantly lower prediction accuracy for the PURPOSIVE tree models relative to the DESIGN models, indicating that non-probabilistic sample surveys may generate models with limited predictive capability. These differences were consistent across all four lichen species, with 11 of the 12 possible species and sample survey type comparisons having significantly lower accuracy rates. Some differences in accuracy were as large as 50%. The classification tree structures also differed considerably both among and within the modelled species, depending on the sample survey form. Overlap in the predictor variables selected by the DESIGN and PURPOSIVE tree models ranged from only 20% to 38%, indicating the classification trees fit the two evaluated survey forms on different sets of predictor variables. The magnitude of these differences in predictor variables throws doubt on ecological interpretation derived from prediction models based on non-probabilistic sample surveys.     

What do consumers believe about future gasoline prices?

A full understanding of how gasoline prices affect consumer behavior frequently requires information on how consumers forecast future gasoline prices. We provide the first evidence on the nature of these forecasts by analyzing two decades of data on gasoline price expectations from the Michigan Survey of Consumers. We find that average consumer beliefs are typically indistinguishable from a no-change forecast, justifying an assumption commonly made in the literature on consumer valuation of energy efficiency. We also provide evidence on circumstances in which consumer forecasts are likely to deviate from no-change and on significant cross-consumer forecast heterogeneity.     

Combining a generic process-based productivity model and a statistical classification method to predict the presence and absence of tree species in the Pacific Northwest,U.S.A.

Although long-lived tree species experience considerable environmental variation over their life spans, their geographical distributions reflect sensitivity mainly to mean monthly climatic conditions. We introduce an approach that incorporates a physiologically based growth model to illustrate how a half-dozen tree species differ in their responses to monthly variation in four climatic-related variables: water availability, deviations from an optimum temperature, atmospheric humidity deficits, and the frequency of frost. Rather than use climatic data directly to correlate with a species’ distribution, we assess the relative constraints of each of the four variables as they affect predicted monthly photosynthesis for Douglas-fir, the most widely distributed species in the region. We apply an automated regression-tree analysis to create a suite of rules, which differentially rank the relative importance of the four climatic modifiers for each species, and provide a basis for predicting a species’ presence or absence on 3737 uniformly distributed U.S. Forest Services’ Forest Inventory and Analysis (FIA) field survey plots. Results of this generalized rule-based approach were encouraging, with weighted accuracy, which combines the correct prediction of both presence and absence on FIA survey plots, averaging 87%. A wider sampling of climatic conditions throughout the full range of a species’ distribution should improve the basis for creating rules and the possibility of predicting future shifts in the geographic distribution of species.     

Conducting conservation social science surveys online

The coronavirus (COVID-19) pandemic is affecting the environment and conservation research in fundamental ways. Many conservation social scientists are now administering survey questionnaires online, but they must do so while ensuring rigor in data collection. Further, they must address a suite of unique challenges, such as the increasing use of mobile devices by participants and avoiding bots or other survey fraud. We reviewed recent literature on online survey methods to examine the state of the field related to online data collection and dissemination. We illustrate the review with examples of key methodological decisions made during a recent national study of people who feed wild birds, in which survey respondents were recruited through an online panel and a sample generated via a project participant list. Conducting surveys online affords new opportunities for participant recruitment, design, and pilot testing. For instance, online survey panels can provide quick access to large and diverse samples of people. Based on the literature review and our own experiences, we suggest that to ensure high-quality online surveys one should account for potential sampling and nonresponse error, design survey instruments for use on multiple devices, test the instrument, and use multiple protocols to identify data quality problems. We also suggest that research funders, journal editors, and policy makers can all play a role in ensuring high-quality survey data are used to inform effective conservation programs and policies.     

Ability of Matrix Models to Explain the Past and Predict the Future of Plant Populations

Uncertainty associated with ecological forecasts has long been recognized, but forecast accuracy is rarely quantified. We evaluated how well data on 82 populations of 20 species of plants spanning 3 continents explained and predicted plant population dynamics. We parameterized stage‐based matrix models with demographic data from individually marked plants and determined how well these models forecast population sizes observed at least 5 years into the future. Simple demographic models forecasted population dynamics poorly; only 40% of observed population sizes fell within our forecasts’ 95% confidence limits. However, these models explained population dynamics during the years in which data were collected; observed changes in population size during the data‐collection period were strongly positively correlated with population growth rate. Thus, these models are at least a sound way to quantify population status. Poor forecasts were not associated with the number of individual plants or years of data. We tested whether vital rates were density dependent and found both positive and negative density dependence. However, density dependence was not associated with forecast error. Forecast error was significantly associated with environmental differences between the data collection and forecast periods. To forecast population fates, more detailed models, such as those that project how environments are likely to change and how these changes will affect population dynamics, may be needed. Such detailed models are not always feasible. Thus, it may be wiser to make risk‐averse decisions than to expect precise forecasts from models. Habilidad de los Modelos Matriciales para Explicar el Pasado y Predecir el Futuro de las Poblaciones de Plantas     

A constrained least-squares approach to combine bottom-up and top-down CO2 flux estimates

Terrestrial CO₂ flux estimates are obtained from two fundamentally different methods generally termed bottom-up and top-down approaches. Inventory methods are one type of bottom-up approach which uses various sources of information such as crop production surveys and forest monitoring data to estimate the annual CO₂ flux at locations covering a study region. Top-down approaches are various types of atmospheric inversion methods which use CO₂ concentration measurements from monitoring towers and atmospheric transport models to estimate CO₂ flux over a study region. Both methods can also quantify the uncertainty associated with their estimates. Historically, these two approaches have produced estimates that differ considerably. The goal of this work is to construct a statistical model which sensibly combines estimates from the two approaches to produce a new estimate of CO₂ flux for our study region. The two approaches have complementary strengths and weaknesses, and our results show that certain aspects of the uncertainty associated with each of the approaches are greatly reduced by combining the methods. Our model is purposefully simple and designed to take the two approaches’ estimates and measures of uncertainty at ‘face value’. Specifically, we use a constrained least-squares approach to appropriately weigh the estimates by the inverse of their variance, and the constraint imposes agreement between the two sources. Our application involves nearly 18,000 flux estimates for the upper midwest United States. The constrained dependencies result in a non-sparse covariance matrix, but computation requires only minutes due to the structure of the model.     

Choosing a Management Strategy: Two Structured Decision-Making Methods for Evaluating the Predictions of Stochastic Simulation Models

We describe two structured decision-making methods—one using a hierarchy of goals and a second using ranking on the sum of weighted criteria—that may be useful for many practical conservation problems, particularly when advisory groups evaluate the output of simulation models. We illustrate both methods by applying them to the problem of choosing a management strategy to address the "mobbing" problem in endangered Hawaiian monk seals. Both methods require estimates of the probabilities of various outcomes, such as a population size of more than 400 seals after 20 years under a specific management regime. We used a simulation model of a small monk seal population to generate these probabilities. Both methods provide an explicit, well-documented, and reproducible decision process that helps justify the decision. Furthermore, they are easy for those untrained in decision analysis to understand and use, they focus discussion on management objectives, they facilitate an examination of trade-offs in the light of multiple and sometimes conflicting objectives, they are suitable for use in workshops, and, at least in our example, they lead to management recommendations that are not highly sensitive to minor changes in probability estimates or other factors.     

A framework for adapting survey design through time for wildlife population assessment

Sampling strategies for monitoring the status and trends in wildlife populations are often determined before the first survey is undertaken. However, there may be little information about the distribution of the population and so the sample design may be inefficient. Through time, as data are collected, more information about the distribution of animals in the survey region is obtained but it can be difficult to incorporate this information in the survey design. This paper introduces a framework for monitoring motile wildlife populations within which the design of future surveys can be adapted using data from past surveys whilst ensuring consistency in design-based estimates of status and trends through time. In each survey, part of the sample is selected from the previous survey sample using simple random sampling. The rest is selected with inclusion probability proportional to predicted abundance. Abundance is predicted using a model constructed from previous survey data and covariates for the whole survey region. Unbiased design-based estimators of status and trends and their variances are derived from two-phase sampling theory. Simulations over the short and long-term indicate that in general more precise estimates of status and trends are obtained using this mixed strategy than a strategy in which all of the sample is retained or all selected with probability proportional to predicted abundance. Furthermore the mixed strategy is robust to poor predictions of abundance. Estimates of status are more precise than those obtained from a rotating panel design.     

Static modeling of dynamic recreation behavior: Implications for prediction and welfare estimation

This paper examines the consequences of using a static model of recreation trip-taking behavior when the underlying decision problem is dynamic. Specifically we examine the implications for trip forecasting and welfare estimation using a panel dataset of Lake Michigan salmon anglers for the 1996 and 1997 fishing seasons. We derive and estimate both a structural dynamic model using Bellman's equation, and a reduced-form static model with trip probability expressions mimicking those of the dynamic model. We illustrate an inherent identification problem in the reduced-form model that creates biased welfare estimates, and we discuss the general implications of this for the interpretation of preference parameters in static models. We then use both models to simulate trip taking behavior and show that although their in-sample trip forecasts are similar, their welfare estimates and out-of-sample forecasts are quite different.     

A comparison of two benthic survey methods

The belt transect and the flowmeter methods, used to survey sedentary benthic organisms, were compared based on abundance estimates of tridacnid clams in the Cagayan Islands, Philippines, in April 1985. Two-way ANOVA and chi-square tests (P<0.05) show that both methods recorded similar estimates for the number of clams (regardless of species), both methods gave similar estimates for the number of individuals per species, and one or both methods may be biased for certain sizes of clams. Individuals smaller than 6 cm shell length seemed to be underestimated by the flowmeter method. Future studies must investigate the size selectivity of the method used, for instance by conducting permanent belt, transect surveys of varied belt widths (e.g. 1, 1.5,2,2.5 m) and then comparing the recorded lengths of the most abundant clams. This will aid investigators to evaluate their data properly, and enable comparison of clam-stock estimates between surveys.     

Application of Conservation Biology Research to Management

Abstract: We conducted a survey of all primary authors of "Contributed Papers" and "Research Notes" in Conservation Biology from 1987 to 1998 to assess the perceived effectiveness of published management recommendations. No systematic survey has previously assessed the degree to which authors believe that resource managers are using the growing body of research published in Conservation Biology . In March 1999, we sent surveys to 667 authors of 790 published papers, asking whether their papers included management recommendations, whether such recommendations have been used in practice, and why they believed they have or have not been used. We received completed surveys from 198 authors of 223 papers. The percentage of papers that included management recommendations increased from 1987 to 1991, then stabilized at about 75%. Author perception of the use of management recommendations generally increased over the 5 years from 1994 to 1998. Initiatives of federal, state, and local agencies were the most often-cited cause of successful implementation of management recommendations, accounting for over half of all such cases. Our survey suggests that authors in Conservation Biology have increased the use of explicit management recommendations in their papers and that authors believe their recommendations are being used to an increasing degree.     

Evaluating the performance of methods for estimating the abundance of rapidly declining coastal shark populations

Accurately surveying shark populations is critical to monitoring precipitous ongoing declines in shark abundance and interpreting the effects that these reductions are having on ecosystems. To evaluate the effectiveness of existing survey tools, we used field trials and computer simulations to critically examine the operation of four common methods for counting coastal sharks: stationary point counts, belt transects, video surveys, and mark and recapture abundance estimators. Empirical and theoretical results suggest that (1) survey method selection has a strong impact on the estimates of shark density that are produced, (2) standardizations by survey duration are needed to properly interpret and compare survey outputs, (3) increasing survey size does not necessarily increase survey precision, and (4) methods that yield the highest density estimates are not always the most accurate. These findings challenge some of the assumptions traditionally associated with surveying mobile marine animals. Of the methods we trialed, 8 x 50 m belt transects and a 20 m radius point count produced the most accurate estimates of shark density. These findings can help to improve the ways we monitor, manage, and understand the ecology of globally imperiled coastal shark populations.     

Addressing onsite sampling in recreation site choice models

Independent experts and politicians have criticized statistical analyses of recreation behavior, which rely upon onsite samples due to their potential for biased inference. The use of onsite sampling usually reflects data or budgetary constraints, but can lead to two primary forms of bias in site choice models. First, the strategy entails sampling site choices rather than sampling individuals—a form of bias called endogenous stratification. Under these conditions, sample choices may not reflect the site choices of the true population. Second, exogenous attributes of the individuals sampled onsite may differ from the attributes of individuals in the population—the most common form in recreation demand is avidity bias. We propose addressing these biases by combining two the existing methods: Weighted Exogenous Stratification Maximum Likelihood estimation and propensity score estimation. We use the National Marine Fisheries Service's Marine Recreational Fishing Statistics Survey to illustrate methods of bias reduction, employing both simulated and empirical applications. We find that propensity score based weights can significantly reduce bias in estimation. Our results indicate that failure to account for these biases can overstate anglers' willingness to pay for improvements in fishing catch, but weighted models exhibit higher variance of parameter estimates and willingness to pay.     

Bridging information domains to improve ecological understanding of biological invasions in a marine ecosystem

We devised a practical method for integrating information on 2 marine invasive species using 3 different approaches: standardized ecological monitoring, online-reporting databases, and surveys of anglers and crabbers. Focusing on 2 recently introduced species with different characteristics, the Asian shore crab (Hemigrapsus sanguineus) and Chinese mitten crab (Eriocheir sinensis), in the Hudson-Raritan watershed of New York and New Jersey, we used sensitivity analyses to explore the relative contribution of each information source to knowledge of species abundance and distribution. All 3 information sources contributed something unique to understanding abundance and distribution of the introduced crabs. Online and survey data on Asian shore crabs significantly affected predictions of abundance, whereas monitoring data did not. When survey data were omitted, abundance estimates were unchanged over time, but when they were included, the model predicted an increased abundance in 2012. All 3 data sets for the Asian shore crab significantly affected estimates of species coverage; surveys had the biggest influence, increasing range size by 4097.25 km². For the catadromous Chinese mitten crab, ecological monitoring data collected in freshwater shortly after the original sighting significantly shaped model estimates for abundance and documented the establishment phase of the mitten crab in an area outside the spatial scope of the surveyed resource users. However, the survey data significantly enlarged mitten crab range-size estimates by 6498.01 km². By demonstrating that data integration produced an image of the invasion process that would not have emerged had we used any 1 method individually, model results provide evidence for the advantages of an interdisciplinary approach.