Jarrett J. Barber

Quantifying ecological memory in plant and ecosystem processes

The role of time in ecology has a long history of investigation, but ecologists have largely restricted their attention to the influence of concurrent abiotic conditions on rates and magnitudes of important ecological processes. Recently, however, ecologists have improved their understanding of ecological processes by explicitly considering the effects of antecedent conditions. To broadly help in studying the role of time, we evaluate the length, temporal pattern, and strength of memory with respect to the influence of antecedent conditions on current ecological dynamics. We developed the stochastic antecedent modelling (SAM) framework as a flexible analytic approach for evaluating exogenous and endogenous process components of memory in a system of interest. We designed SAM to be useful in revealing novel insights promoting further study, illustrated in four examples with different degrees of complexity and varying time scales: stomatal conductance, soil respiration, ecosystem productivity, and tree growth. Models with antecedent effects explained an additional 18-28% of response variation compared to models without antecedent effects. Moreover, SAM also enabled identification of potential mechanisms that underlie components of memory, thus revealing temporal properties that are not apparent from traditional treatments of ecological time-series data and facilitating new hypothesis generation and additional research.

Elk migration patterns and human activity influence wolf habitat use in the Greater Yellowstone Ecosystem

Identifying the ecological dynamics underlying human-wildlife conflicts is important for the management and conservation of wildlife populations. In landscapes still occupied by large carnivores, many ungulate prey species migrate seasonally, yet little empirical research has explored the relationship between carnivore distribution and ungulate migration strategy. In this study, we evaluate the influence of elk (Cervus elaphus) distribution and other landscape features on wolf (Canis lupus) habitat use in an area of chronic wolf-livestock conflict in the Greater Yellowstone Ecosystem, USA. Using three years of fine-scale wolf (n = 14) and elk (n = 81) movement data, we compared the seasonal habitat use of wolves in an area dominated by migratory elk with that of wolves in an adjacent area dominated by resident elk. Most migratory elk vacate the associated winter wolf territories each summer via a 40-60 km migration, whereas resident elk remain accessible to wolves year-round. We used a generalized linear model to compare the relative probability of wolf use as a function of GIS-based habitat covariates in the migratory and resident elk areas. Although wolves in both areas used elk-rich habitat all year, elk density in summer had a weaker influence on the habitat use of wolves in the migratory elk area than the resident elk area. Wolves employed a number of alternative strategies to cope with the departure of migratory elk. Wolves in the two areas also differed in their disposition toward roads. In winter, wolves in the migratory elk area used habitat close to roads, while wolves in the resident elk area avoided roads. In summer, wolves in the migratory elk area were indifferent to roads, while wolves in resident elk areas strongly avoided roads, presumably due to the location of dens and summering elk combined with different traffic levels. Study results can help wildlife managers to anticipate the movements and establishment of wolf packs as they expand into areas with migratory or resident prey populations, varying levels of human activity, and front-country rangelands with potential for conflicts with livestock.

Hierarchical spatial modeling for estimation of population size

Estimation of population size has traditionally been viewed from a finite population sampling perspective. Typically, the objective is to obtain an estimate of the total population count of individuals within some region. Often, some stratification scheme is used to estimate counts on subregions, whereby the total count is obtained by aggregation with weights, say, proportional to the areas of the subregions.We offer an alternative to the finite population sampling approach for estimating population size. The method does not require that the subregions on which counts are available form a complete partition of the region of interest. In fact, we envision counts coming from areal units that are small relative to the entire study region and that the total area sampled is a very small proportion of the total study area. In extrapolating to the entire region, we might benefit from assuming that there is spatial structure to the counts. We implement this by modeling the intensity surface as a realization from a spatially correlated random process. In the case of multiple population or species counts, we use the linear model of coregionalization to specify a multivariate process which provides associated intensity surfaces hence association between counts within and across areal units.We illustrate the method of population size estimation with simulated data and with tree counts from a Southwestern pinyon-juniper woodland data set.

Quantifying antecedent climatic drivers of tree growth in the Southwestern US

Summary 1.Variation in antecedent (past) climate conditions is likely to govern tree growth over long periods of time. Antecedent conditions are rarely considered in models of tree growth, representing a weakness in quantitative understanding of forest responses to climate variations. 2.We applied the stochastic antecedent modelling (SAM) framework to 367 International Tree Ring Data Bank (ITRDB) chronologies in the southwestern US (‘Southwest’) representing eight conifer species. To better understand climatic and physiologic controls on tree growth, we quantify the effects of antecedent precipitation, temperature, and Palmer Drought Severity Index (PDSI) over 60 months preceding and including the year of ring formation. 3.In Pinus edulis, P. ponderosa, and Pseudotsuga menziesii, growth responded primarily to recent precipitation and temperature conditions (43-49% of the response was driven by conditions during the year of ring formation), but to less recent PDSI conditions (>50% of response driven by conditions 13-48 months prior to the year of ring formation), though PDSI significantly affected growth at only 21% of sites. Combining extensive tree-ring data with monthly resolution climate data also reveals key climatic events, such as the effect of monsoon arrival date on growth, especially in P. menziesii, highlighting the ability of the SAM framework to identify climate effects at multiple time scales. 4.Sensitivity to antecedent climate, baseline growth at average climate conditions, and the strength of 1st order autocorrelation varied spatially, suggesting variation in mean growing conditions, non-structural carbohydrate storage, and/or seasonal precipitation contribution of the North American Monsoon may drive differences in growth sensitivities across species’ ranges. 5.Synthesis - Our findings provide further evidence for multi-year legacy effects of climate conditions, particularly drought metrics, on tree growth. Antecedent climate and especially drought are key drivers of growth in these species, and associated climatic sensitivities and growth indices vary spatially. We argue such factors should be considered in modelling efforts. The spatial variability in antecedent climate sensitivities points to key differences in how different populations within a species range may respond to climate change, particularly if timing of weather events, such as monsoon arrival date, or annual precipitation amounts undergo significant changes. This article is protected by copyright. All rights reserved.

Plant and Ecosystem Memory

Of course plants do not have brains and, thus, cannot actually remember what happened to them in the past. Although plants cannot remember, however, we use “memory” as a metaphor to refer to the effect of the past on current and future plant and ecosystem functioning. Such memory effects have been repeatedly shown in ecosystems (such as deserts) that are often defined by highly variable environmental conditions, where air temperatures, humidity, and soil water availability can differ greatly from one day to the next, week to week, among seasons, and year to year. For example, the amount of carbon that an ecosystem “releases” to the atmosphere through plant respiration and microbial decomposition of organic matter (e.g., dead plant leaves and roots) is strongly controlled by past (antecedent) temperature and soil water content. Similarly, annual tree growth, as recorded in tree-ring widths (see Figure 2), is strongly controlled by temperature conditions and precipitation received months and years before the initiation of new growth. On shorter time scales, the rate at which a plant leaf takes up CO2 (via photosynthesis) and releases water vapor (via transpiration) to the atmosphere is regulated by past humidity, temperature, and soil moisture availability. Thus, plants and ecosystems do remember. That is, past environmental conditions are imprinted on current and future plant and ecosystem functions. Of course, memories of our past are important for who we are and for what we will become, and ecologists have recognized their potential for understanding and predicting plant and ecosystem behavior.

Modeling unobserved true position using multiple sources and information semantics

We present a modeling framework that supports semantically informed statistical inference about unobserved true location and positional uncertainty for geographic information spanning multiple sources. We demonstrate the use of a semantic representation of information sources to support construction of a Bayesian belief network that operationalizes the data integration process. Our approach allows for positional data, metadata, and other ancillary and derived information to inform inference regarding unobserved true position. In our application, we use two line feature datasets and a set of GPS data points describing a portion of the street network in Laramie, Wyoming. Using source metadata we inform prior distributions. Additionally, we use feature straightness to illustrate how form and process – gridded streets and the process of the Public Land Survey – can be used to improve inference for true position. The presented modeling framework is suitable for multiple data sources when the best data are not necessarily known and when the information semantics associated with the input data can be described in a systematic way.

Estimating the Hatchery Fraction of a Natural Population: A Bayesian Approach

Abstract There is strong and growing interest in estimating the proportion of hatchery fish that are in a natural population (the hatchery fraction). In a sample of fish from the relevant population, some are observed to be marked, indicating their origin as hatchery fish. The observed proportion of marked fish is usually less than the actual hatchery fraction, since the observed proportion is determined by the proportion originally marked, differential survival (usually lower) of marked fish relative to unmarked hatchery fish, and rates of mark retention and detection. Bayesian methods can work well in a setting such as this, in which empirical data are limited but for which there may be considerable expert judgment regarding these values. We explored a Bayesian estimation of the hatchery fraction using Monte Carlo–Markov chain methods. Based on our findings, we created an interactive Excel tool to implement the algorithm, which we have made available for free. Received March 30, 2011; accepted June 21, 2011

Using Rubber Sheets To Infer True Map Location

Recent technological advances now facilitate the collection of and access to an enormous amount of spatial data in the form of electronic maps, and geographic information systems (GIS) are a popular and important tool used to store, manipulate, and analyze such data. But, the apparent sophistication with which electronic maps are depicted conveys a sense of accuracy that is often unwarranted for many uses. Various aspects of a map are susceptible to inaccuracies, including the position (i.e., numerical coordinates) of map features, values attributed to a map feature—is it a primary road or secondary road, a spruce forest or pine forest—and even the unintentional omission of features altogether. Such considerations fall within the realm of what is generally referred to as “map accuracy,” and studies of how map errors propagate through various GIS operations are common—socalled error propagation studies. But, relative to the existing volume of spatial data and the ease with which data is produced, these methods seem comparably underdeveloped, and, somewhat curiously, statisticians on the whole—in particular, spatial statisticians—do not seem to be playing much of a role in such investigations. After all, statisticians are somehow the collective authority on uncertainty, and spatial statisticians, you might think, should know something about map uncertainty. Here, we investigate only one of the many aspects of map accuracy—often referred to as map “positional accuracy,” “positional uncertainty,” or “positional error”—and we offer an initial perspective to statisticians in the form of a statistical model for positional uncertainty that allows us to infer features’ true positions.

Combining and comparing multiple serial dilution assays of particles in solution: application to brucellosis in elk of the Greater Yellowstone Ecosystem

The concentration detection threshold (CDT) is the concentration of particles in solution beyond which a (serial dilution) assay detects particle presence. By our account, CDTs typically are not estimated but are fixed at some value. Setting a CDT to zero (