A5018991559- Marine and fisheries research
- Fish Ecology and Management Studies
- Marine Bivalve and Aquaculture Studies
- Isotope Analysis in Ecology
- Reproductive biology and impacts on aquatic species
- Identification and Quantification in Food
- Species Distribution and Climate Change
- Aquaculture Nutrition and Growth
- Statistical Methods and Bayesian Inference
- Ecology and Vegetation Dynamics Studies
- Environmental DNA in Biodiversity Studies
- Coral and Marine Ecosystems Studies
- Ichthyology and Marine Biology
- Advanced Polymer Synthesis and Characterization
- Bayesian Modeling and Causal Inference
- Data Analysis with R
- Oceanographic and Atmospheric Processes
- Genetic and phenotypic traits in livestock
- Marine animal studies overview
- Gaussian Processes and Bayesian Inference
- Mathematical and Theoretical Epidemiology and Ecology Models
- Ecosystem dynamics and resilience
- Archaeological and Historical Studies
- Simulation Techniques and Applications
- Synthetic Organic Chemistry Methods
Technical University of Denmark
2014-2024
University of Zurich
2017
McMaster University
2017
University of Bergen
2017
ETH Zurich
2017
International Council for the Exploration of the Sea
2017
Danish Centre for Marine Research
2003-2015
University of Hawaiʻi at Mānoa
2006-2009
University of Hawaii System
2006-2009
Joint Institute for Nuclear Research
2009
Count data can be analyzed using generalized linear mixed models when observations are correlated in ways that require random effects.However, count often zero-inflated, containing more zeros than would expected from the typical error distributions.We present a new package, glmmTMB, and compare it to other R packages fit zero-inflated models.The glmmTMB package fits many types of GLMMs extensions, including with continuously distributed responses, but here we focus on responses.glmmTMB is...
Many criteria for statistical parameter estimation, such as maximum likelihood, are formulated a nonlinear optimization problem. Automatic Differentiation Model Builder (ADMB) is programming framework based on automatic differentiation, aimed at highly models with large number of parameters. The benefits using AD computational efficiency and high numerical accuracy, both crucial in many practical problems. We describe the basic components underlying philosophy ADMB, an emphasis functionality...
TMB is an open source R package that enables quick implementation of complex nonlinear random effects (latent variable) models in a manner similar to the established AD Model Builder (ADMB, http://admb-project.org/; Fournier et al. 2011). In addition, it offers easy access parallel computations. The user defines joint likelihood for data and as C++ template function, while all other operations are done R; e.g., reading data. evaluates maximizes Laplace approximation marginal where...
Abstract Ecological phenomena are often measured in the form of count data. These data can be analyzed using generalized linear mixed models (GLMMs) when observations correlated ways that require random effects. However, zero-inflated , containing more zeros than would expected from standard error distributions used GLMMs, e.g., parasite counts may exactly zero for hosts with effective immune defenses but vary according to a negative binomial distribution non-resistant hosts. We present new...
State-space models (SSMs) are an important modeling framework for analyzing ecological time series. These hierarchical commonly used to model population dynamics, animal movement, and capture-recapture data, now increasingly being other processes. SSMs popular because they flexible the natural variation in processes separately from observation error. Their flexibility allows ecologists continuous, count, binary, categorical data with linear or nonlinear that evolve discrete continuous time....
Worldwide a number of fish stocks have collapsed because overfishing and climate-induced ecosystem changes. Developing ecosystem-based fisheries management (EBFM) to prevent these catastrophic events in the future requires ecological models incorporating both internal food-web dynamics external drivers such as fishing climate. Using stochastic model for large marine (i.e., Baltic Sea) hosting commercially important cod stock, we were able reconstruct history stock. Moreover demonstrate that...
Summary Ecologists often use nonlinear fitting techniques to estimate the parameters of complex ecological models, with attendant frustration. This paper compares three open‐source model tools and discusses general strategies for defining models. R is convenient (relatively) easy learn, AD M odel B uilder fast robust but comes a steep learning curve, while BUGS provides greatest flexibility at price speed. Our model‐fitting suggestions range from cultural advice (where possible, models that...
Abstract An approach to integrate sea surface temperature (SST) measurements into estimates of geolocations calculated by changes in ambient light level from data downloaded pop‐up satellite archival tags (PSAT) is presented. The model an extension based on Kalman filter estimation a state‐space model. uses longitude and latitude estimated light, SST. extra information SST included consistent manner within the milieu filter. technique was evaluated attaching PSATs directly...
Good decision making for fisheries and marine ecosystems requires a capacity to anticipate the consequences of management under different scenarios climate change. The necessary ecological forecasting calls ecosystem-based models capable integrating multiple drivers across trophic levels properly including uncertainty. methodology presented here assesses combined impacts fishing on food-web dynamics provides estimates confidence envelope forecasts. It is applied cod ( Gadus morhua ) in...
Abstract The World Conference on Stock Assessment Methods (July 2013) included a workshop testing assessment methods through simulations. exercise was made up of two steps applied to datasets from 14 representative fish stocks around the world. Step 1 involved applying stock assessments with varying degrees effort dedicated optimizing fit. 2 subset and characteristics given model fits being used generate pseudo-data error. These were then provided modellers consistency checks within...
Fisheries science is concerned with the management and understanding of raising harvesting fish. Fish stocks are assessed using biological fisheries data goal estimating either their total population or biomass. Stock assessment models also make it possible to predict how will respond varying levels fishing pressure in future. Such tools essential overfishing now reducing employment worldwide, turn many serious social, economic, environmental implications. Increasingly, a state-space...
A coherent model is presented to estimate the most probable track of geographic positions directly from a series light measurements. The estimates two per day, without reducing daily data threshold crossing times, its covariance structure designed handle high correlations due for instance local weather conditions, and it can yearly pattern in latitudinal precision by propagating uncertainties through geolocation process. applied one mooring study, GPS drifter buoy numerous simulated cases....
State-space models (SSM) are often used for analyzing complex ecological processes that not observed directly, such as marine animal movement. When outliers present in the measurements, special care is needed analysis to obtain reliable location and process estimates. Here we recommend using Laplace approximation combined with automatic differentiation (as implemented novel R package Template Model Builder; TMB) fast fitting of continuous-time multivariate non-Gaussian SSMs. Through Argos...
Abstract Fish stock assessment models often rely on size- or age-specific observations that are assumed to be statistically independent of each other. In reality, these not raw observations, but rather they estimates from a catch-standardization model similar summary statistics based many fishing hauls and subsamples the size age composition data. Although aggregation mitigates strong intra-haul correlation between sizes/ages is usually found in haul-by-haul data, violations independence...
Fishing and climate change impact the demography of marine fishes, but it is generally ignored that many species are made up genetically distinct locally adapted populations may show idiosyncratic responses to environmental anthropogenic pressures. Here, we track 80 years Atlantic cod (Gadus morhua) population dynamics in West Greenland using DNA from archived otoliths combination with fish niche based modeling. We document how interacting effects high fishing pressure lead dramatic...