Catherine A. Langtimm

ESTIMATING SITE OCCUPANCY RATES WHEN DETECTION PROBABILITIES ARE LESS THAN ONE

Nondetection of a species at a site does not imply that the species is absent unless the probability of detection is 1. We propose a model and likelihood-based method for estimating site occupancy rates when detection probabilities are 0.3). We estimated site occupancy rates for two anuran species at 32 wetland sites in Maryland, USA, from data collected during 2000 as part of an amphibian monitoring program, Frogwatch USA. Site occupancy rates were estimated as 0.49 for American toads (Bufo americanus), a 44% increase over the proportion of sites at which they were actually observed, and as 0.85 for spring peepers (Pseudacris crucifer), slightly above the observed proportion of 0.83.

ESTIMATES OF ANNUAL SURVIVAL PROBABILITIES FOR ADULT FLORIDA MANATEES (TRICHECHUS MANATUS LATIROSTRIS)

The population dynamics of large, long-lived mammals are particularly sensitive to changes in adult survival. Understanding factors affecting survival patterns is therefore critical for developing and testing theories of population dynamics and for developing management strategies aimed at preventing declines or extinction in such taxa. Few studies have used modern analytical approaches for analyzing variation and testing hypotheses about survival probabilities in large mammals. This paper reports a detailed analysis of annual adult survival in the Florida manatee (Trichechus manatus latirostris), an endangered marine mammal, based on a mark–recapture approach. Natural and boat-inflicted scars distinctively “marked” individual manatees that were cataloged in a computer-based photographic system. Photo-documented resightings provided “recaptures.” Using open population models, annual adult-survival probabilities were estimated for manatees observed in winter in three areas of Florida: Blue Spring, Crystal ...

Combining dead recovery, auxiliary observations and robust design data to estimate demographic parameters from marked individuals

Summary When estimating demographic parameters for wild populations, using multiple data sources can increase robustness through greater precision, reducing bias and permitting the estimation of otherwise confounded parameters. We present a method that combines recapture data from marked individuals, collected at a single study site, under a robust design framework, with dead recoveries and auxiliary resightings collected at any time and place. This model permits the joint modelling of survival, permanent and temporary emigration from the study area. We demonstrate that the usefulness of this model is compelling in the case of long-lived species with substantial rates of temporary emigration, to mitigate bias in survival at the end of the time series and to permit conservation decisions based on more current information. We use the case of Florida manatees as an example. Our model can easily be extended to account for an arbitrary number of phenotypic states and account for state uncertainty. The increase in precision overall in vital rates, and the mitigation of bias in survival estimation in the final years of a time series, permits managers to base resource decisions on more robust and timely information. The model also provides the ability to adapt monitoring to changing conditions or specific management objectives, via dynamic allocation of effort to auxiliary resightings.

Estimating parameters of hidden Markov models based on marked individuals: use of robust design data

Development and use of multistate mark-recapture models, which provide estimates of parameters of Markov processes in the face of imperfect detection, have become common over the last 20 years. Recently, estimating parameters of hidden Markov models, where the state of an individual can be uncertain even when it is detected, has received attention. Previous work has shown that ignoring state uncertainty biases estimates of survival and state transition probabilities, thereby reducing the power to detect effects. Efforts to adjust for state uncertainty have included special cases and a general framework for a single sample per period of interest. We provide a flexible framework for adjusting for state uncertainty in multistate models, while utilizing multiple sampling occasions per period of interest to increase precision and remove parameter redundancy. These models also produce direct estimates of state structure for each primary period, even for the case where there is just one sampling occasion. We apply our model to expected-value data, and to data from a study of Florida manatees, to provide examples of the improvement in precision due to secondary capture occasions. We have also implemented these models in program MARK. This general framework could also be used by practitioners to consider constrained models of particular interest, or to model the relationship between within-primary-period parameters (e.g., state structure) and between-primary-period parameters (e.g., state transition probabilities).

New aerial survey and hierarchical model to estimate manatee abundance

ABSTRACT Monitoring the response of endangered and protected species to hydrological restoration is a major component of the adaptive management framework of the Comprehensive Everglades Restoration Plan. The endangered Florida manatee (Trichechus manatus latirostris) lives at the marine-freshwater interface in southwest Florida and is likely to be affected by hydrologic restoration. To provide managers with prerestoration information on distribution and abundance for postrestoration comparison, we developed and implemented a new aerial survey design and hierarchical statistical model to estimate and map abundance of manatees as a function of patch-specific habitat characteristics, indicative of manatee requirements for offshore forage (seagrass), inland fresh drinking water, and warm-water winter refuge. We estimated the number of groups of manatees from dual-observer counts and estimated the number of individuals within groups by removal sampling. Our model is unique in that we jointly analyzed group and individual counts using assumptions that allow probabilities of group detection to depend on group size. Ours is the first analysis of manatee aerial surveys to model spatial and temporal abundance of manatees in association with habitat type while accounting for imperfect detection. We conducted the study in the Ten Thousand Islands area of southwestern Florida, USA, which was expected to be affected by the Picayune Strand Restoration Project to restore hydrology altered for a failed real-estate development. We conducted 11 surveys in 2006, spanning the cold, dry season and warm, wet season. To examine short-term and seasonal changes in distribution we flew paired surveys 1–2 days apart within a given month during the year. Manatees were sparsely distributed across the landscape in small groups. Probability of detection of a group increased with group size; the magnitude of the relationship between group size and detection probability varied among surveys. Probability of detection of individual manatees within a group also differed among surveys, ranging from a low of 0.27 on 11 January to a high of 0.73 on 8 August. During winter surveys, abundance was always higher inland at Port of the Islands (POI), a manatee warm-water aggregation site, than in the other habitat types. During warm-season surveys, highest abundances were estimated in offshore habitat where manatees forage on seagrass. Manatees continued to use POI in summer, but in lower numbers than in winter, possibly to drink freshwater. Abundance in other inland systems and inshore bays was low compared to POI in winter and summer, possibly because of low availability of freshwater. During cold weather, maps of patch abundance of paired surveys showed daily changes in manatee distribution associated with rapid changes in air and water temperature as manatees sought warm water with falling temperatures and seagrass areas with increasing temperatures. Within a habitat type, some patches had higher manatee abundance suggesting differences in quality, possibly due to freshwater flow. If hydrological restoration alters the location of quality habitat, postrestoration comparisons using our methods will document how manatees adjust to new resources, providing managers with information on spatial needs for further monitoring or management. Total abundance for the entire area was similar among survey dates. Credible intervals however were large on a few surveys, and may limit our ability to statistically detect trends in total abundance. Additional modeling of abundance with time- and patch-specific covariates of salinity, water temperature, and seagrass abundance will directly link manatee abundance with physical and biological changes due to restoration and should decrease uncertainty of estimates.

Possible effects of the 2004 and 2005 hurricanes on manatee survival rates and movement

Prior research on manatee (Trichechus manatus latirostris) survival in northwest Florida, based on mark-resighting photo-identification data from 1982–1998, showed that annual adult apparent survival rate was significantly lower during years with extreme storms. Mechanisms that we proposed could have led to lower estimates included stranding, injury from debris, being fatally swept out to sea, or displacement into poorly monitored areas due to storm-generated longshore currents or storm-related loss of habitat. In 2004 and 2005, seven major hurricanes impacted areas of Florida encompassing three regional manatee subpopulations, enabling us to further examine some of these mechanisms. Data from a group of manatees tracked in southwest Florida with satellite transmitters during Hurricanes Charley, Katrina, and Wilma showed that these animals made no significant movement before and during storm passage. Mark-resighting data are being collected to determine if survival rates were lower with the 2004 and 2005 storms.

Non-random Temporary Emigration and the Robust Design: Conditions for Bias at the End of a Time Series

Deviations from model assumptions in the application of capture–recapture models to real life situations can introduce unknown bias. Understanding the type and magnitude of bias under these conditions is important to interpreting model results. In a robust design analysis of long-term photo-documented sighting histories of the endangered Florida manatee, I found high survival rates, high rates of non-random temporary emigration, significant time-dependence, and a diversity of factors affecting temporary emigration that made it difficult to model emigration in any meaningful fashion. Examination of the time-dependent survival estimates indicated a suspicious drop in survival rates near the end of the time series that persisted when the original capture histories were truncated and reanalyzed under a shorter time frame. Given the wide swings in manatee emigration estimates from year to year, a likely source of bias in survival was the convention to resolve confounding of the last survival probability in a time-dependent model with the last emigration probabilities by setting the last unmeasurable emigration probability equal to the previous year’s probability when the equality was actually false. Results of a series of simulations demonstrated that if the unmeasurable temporary emigration probabilities in the last time period were not accurately modeled, an estimation model with significant annual variation in survival probabilities and emigration probabilities produced bias in survival estimates at the end of the study or time series being explored. Furthermore, the bias propagated back in time beyond the last two time periods and the number of years affected varied positively with survival and emigration probabilities. Truncating the data to a shorter time frame and reanalyzing demonstrated that with additional years of data surviving temporary emigrants eventually return and are detected, thus in subsequent analysis unbiased estimates are eventually realized.

Evaluation of habitat suitability index models by global sensitivity and uncertainty analyses: a case study for submerged aquatic vegetation

Habitat suitability index (HSI) models are commonly used to predict habitat quality and species distributions and are used to develop biological surveys, assess reserve and management priorities, and anticipate possible change under different management or climate change scenarios. Important management decisions may be based on model results, often without a clear understanding of the level of uncertainty associated with model outputs. We present an integrated methodology to assess the propagation of uncertainty from both inputs and structure of the HSI models on model outputs (uncertainty analysis: UA) and relative importance of uncertain model inputs and their interactions on the model output uncertainty (global sensitivity analysis: GSA). We illustrate the GSA/UA framework using simulated hydrology input data from a hydrodynamic model representing sea level changes and HSI models for two species of submerged aquatic vegetation (SAV) in southwest Everglades National Park: Vallisneria americana (tape grass) and Halodule wrightii (shoal grass). We found considerable spatial variation in uncertainty for both species, but distributions of HSI scores still allowed discrimination of sites with good versus poor conditions. Ranking of input parameter sensitivities also varied spatially for both species, with high habitat quality sites showing higher sensitivity to different parameters than low-quality sites. HSI models may be especially useful when species distribution data are unavailable, providing means of exploiting widely available environmental datasets to model past, current, and future habitat conditions. The GSA/UA approach provides a general method for better understanding HSI model dynamics, the spatial and temporal variation in uncertainties, and the parameters that contribute most to model uncertainty. Including an uncertainty and sensitivity analysis in modeling efforts as part of the decision-making framework will result in better-informed, more robust decisions.

Reducing bias in survival under nonrandom temporary emigration

Despite intensive monitoring, temporary emigration from the sampling area can induce bias severe enough for managers to discard survival parameter estimates toward the terminus of the times series (terminal bias). Under random temporary emigration, unbiased parameters can be estimated with CJS models. However, unmodeled Markovian temporary emigration causes bias in parameter estimates, and an unobservable state is required to model this type of emigration. The robust design is most flexible when modeling temporary emigration, and partial solutions to mitigate bias have been identified; nonetheless, there are conditions were terminal bias prevails. Long-lived species with high adult survival and highly variable nonrandom temporary emigration present terminal bias in survival estimates, despite being modeled with the robust design and suggested constraints. Because this bias is due to uncertainty about the fate of individuals that are undetected toward the end of the time series, solutions should involve using additional information on survival status or location of these individuals at that time. Using simulation, we evaluated the performance of models that jointly analyze robust design data and an additional source of ancillary data (predictive covariate on temporary emigration, telemetry, dead recovery, or auxiliary resightings) in reducing terminal bias in survival estimates. The auxiliary resighting and predictive covariate models reduced terminal bias the most. Additional telemetry data were effective at reducing terminal bias only when individuals were tracked for a minimum of two years. High adult survival of long-lived species made the joint model with recovery data ineffective at reducing terminal bias because of small-sample bias. The naive constraint model (last and penultimate temporary emigration parameters made equal), was the least efficient, although still able to reduce terminal bias when compared to an unconstrained model. Joint analysis of several sources of data improved parameter estimates and reduced terminal bias. Efforts to incorporate or acquire such data should be considered by researchers and wildlife managers, especially in the years leading up to status assessments of species of interest. Simulation modeling is a very cost-effective method to explore the potential impacts of using different sources of data to produce high-quality demographic data to inform management.

Numerical computation of hurricane effects on historic coastal hydrology in Southern Florida

IntroductionNumerical models are critical for assessing the effects of sea level rise (SLR), hurricanes, and storm surge on vegetation change in the Everglades National Park. The model must be capable of representing short-timescale hydrodynamics, salinity transport, and groundwater interaction. However, there is also a strong need to adapt these numerical models to hindcast past conditions in order to examine long-term effects on the distribution of vegetation that cannot be determined using only the modern record.MethodsBased on parameters developed for a numerical model developed for the recent 1996 to 2004 period, a hindcast model was developed to represent sea level and water management for the period of 1926 to 1932, constrained by the limited hydrology and meteorology data available from the historic past. Realistic hurricane-wind and storm surge representations, required for the hindcast model, are based on information synthesized from modern storm data. A series of simulation scenarios with various hurricane representations inserted into both hindcast and recent numerical models were used to assess the utility of the storm representation in the model and compare the two simulations.ResultsThe comparison of the hindcast and recent models showed differences in the hydrology patterns that are consistent with known differences in water delivery systems and sea level rise. A 30× lower-resolution spatially variable wind grid for the hindcast produced similar results to the original high-resolution full wind grid representation of the recent simulation. Storm effects on hydrologic patterns demonstrated with the simulations show hydrologic processes that could have a long-term effect on vegetation change.ConclusionsThe hindcast simulation estimated hydrologic processes for the 1926 to 1932 period. It shows promise as a simulator in long-term ecological studies to test hypotheses based on theoretical or empirical-based studies at larger landscape scales.