Prediction of marine species distribution from presence–absence acoustic data: comparing the fitting efficiency and the predictive capacity of conventional and novel distribution models

Hydrobiologia ◽  
2011 ◽  
Vol 670 (1) ◽  
pp. 241-266 ◽  
Author(s):  
A. Palialexis ◽  
S. Georgakarakos ◽  
I. Karakassis ◽  
K. Lika ◽  
V. D. Valavanis
Keyword(s):  
Species Distribution ◽  
Marine Species ◽  
Distribution Models ◽  
Predictive Capacity ◽  
Acoustic Data

Using verified species distribution models to inform the conservation of a rare marine species

2016 ◽  
Vol 22 (7) ◽  
pp. 808-822 ◽  
Author(s):  
David A. Stirling ◽  
Philip Boulcott ◽  
Beth E. Scott ◽  
Peter J. Wright
Keyword(s):  
Species Distribution ◽  
Species Distribution Models ◽  
Marine Species ◽  
Distribution Models

Habitat representativeness score (HRS): a novel concept for objectively assessing the suitability of survey coverage for modelling the distribution of marine species

2010 ◽  
Vol 90 (7) ◽  
pp. 1269-1277 ◽  
Author(s):  
Colin D. MacLeod
Keyword(s):  
Species Distribution ◽  
Environmental Variables ◽  
Survey Design ◽  
Objective Assessment ◽  
Principal Component ◽  
Marine Species ◽  
Representative Survey ◽  
Distribution Models ◽  
Data Set ◽  
Minimum Number

The occurrence of most species is linked to the distribution of specific combinations of environmental variables that define their occupied niche. As a result, the relationship between environmental variables and species occurrence can be used to model species distribution. However, when collecting data to construct such models, it is preferable to ensure that the survey coverage is representative of all available habitat combinations within the area as a whole to ensure that the model does not under- or over-estimate the actual species distribution. By using multi-variate statistical techniques, a habitat representativeness score (HRS) can be calculated to provide an objective assessment of whether a specific survey coverage will collect (or has collected) data that are representative of all available habitat variable combinations in an area. To demonstrate this approach, HRSs calculated using principal component analysis were used to assess the minimum number of evenly-spaced parallel north–south surveys required to adequately survey two study areas with differing levels of environmental heterogeneity for all available combinations of four habitat variables. For the more environmentally homogeneous study area, the HRS suggests that for this survey design a minimum of five evenly-spaced parallel transects, covering around 5% of the study area, would be required to obtain representative survey coverage for these four variables. However, for the more heterogeneous study area, at least eight evenly-spaced parallel transects, covering around 9% of the study area, would be required. Therefore, for a given survey design, more survey effort is required to obtain a representative survey coverage when the survey area is more variable. In both cases, conducting fewer surveys than these minimum values would produce an unrepresentative data set and this could potentially lead to the production of species distribution models that do not accurately reflect the true species distribution.

scholarly journals Jointly modeling marine species to inform the effects of environmental change on an ecological community in the Northwest Atlantic

2022 ◽  
Vol 12 (1) ◽  
Author(s):  
Sarah M. Roberts ◽  
Patrick N. Halpin ◽  
James S. Clark
Keyword(s):  
Marine Fish ◽  
Species Distribution ◽  
Environmental Effects ◽  
Species Distribution Model ◽  
Predictive Performance ◽  
Single Species ◽  
Marine Species ◽  
Joint Modeling ◽  
Distribution Model ◽  
Distribution Models

AbstractSingle species distribution models (SSDMs) are typically used to understand and predict the distribution and abundance of marine fish by fitting distribution models for each species independently to a combination of abiotic environmental variables. However, species abundances and distributions are influenced by abiotic environmental preferences as well as biotic dependencies such as interspecific competition and predation. When species interact, a joint species distribution model (JSDM) will allow for valid inference of environmental effects. We built a joint species distribution model of marine fish and invertebrates of the Northeast US Continental Shelf, providing evidence on species relationships with the environment as well as the likelihood of species to covary. Predictive performance is similar to SSDMs but the Bayesian joint modeling approach provides two main advantages over single species modeling: (1) the JSDM directly estimates the significance of environmental effects; and (2) predicted species richness accounts for species dependencies. An additional value of JSDMs is that the conditional prediction of species distributions can use not only the environmental associations of species, but also the presence and abundance of other species when forecasting future climatic associations.

scholarly journals Incorporating physiology into species distribution models moderates the projected impact of warming on selected Mediterranean marine species

Ecography ◽  
2020 ◽  
Vol 43 (7) ◽  
pp. 1090-1106 ◽  
Author(s):  
Inbal Gamliel ◽  
Yehezkel Buba ◽  
Tamar Guy‐Haim ◽  
Tal Garval ◽  
Demian Willette ◽  
...  
Keyword(s):  
Species Distribution ◽  
Species Distribution Models ◽  
Marine Species ◽  
Distribution Models

scholarly journals A Robust Prediction Model for Species Distribution Using Bagging Ensembles with Deep Neural Networks

2021 ◽  
Vol 13 (8) ◽  
pp. 1495
Author(s):  
Jehyeok Rew ◽  
Yongjang Cho ◽  
Eenjun Hwang
Keyword(s):  
Neural Networks ◽  
Species Distribution ◽  
Best Practice ◽  
Deep Neural Networks ◽  
Species Distribution Models ◽  
Species Distribution Model ◽  
Environmental Data ◽  
Distribution Model ◽  
Distribution Models ◽  
Robust Prediction

Species distribution models have been used for various purposes, such as conserving species, discovering potential habitats, and obtaining evolutionary insights by predicting species occurrence. Many statistical and machine-learning-based approaches have been proposed to construct effective species distribution models, but with limited success due to spatial biases in presences and imbalanced presence-absences. We propose a novel species distribution model to address these problems based on bootstrap aggregating (bagging) ensembles of deep neural networks (DNNs). We first generate bootstraps considering presence-absence data on spatial balance to alleviate the bias problem. Then we construct DNNs using environmental data from presence and absence locations, and finally combine these into an ensemble model using three voting methods to improve prediction accuracy. Extensive experiments verified the proposed model’s effectiveness for species in South Korea using crowdsourced observations that have spatial biases. The proposed model achieved more accurate and robust prediction results than the current best practice models.

Sign in / Sign up

Export Citation Format

Share Document