Prediction of large whale distributions: A comparison of presence-absence and presence-only modeling techniques

Paul C. Fiedler, Jessica V. Redfern, Karin A. Forney, Daniel M. Palacios, Corey Sheredy, Kristin Rasmussen, Ignacio García-Godos, Luis Santillán, Michael J. Tetley, Fernando Félix, Lisa T. Ballance

Resultado de la investigación: Contribución a una revistaArtículoInvestigaciónrevisión exhaustiva

2 Citas (Scopus)

Resumen

© 2018 Fiedler, Redfern, Forney, Palacios, Sheredy, Rasmussen, García-Godos, Santillán, Tetley, Félix and Ballance. Species distribution models that predict species occurrence or density by quantifying relationships with environmental variables are used for a variety of scientific investigations and management applications. For endangered species, such as large whales, models help to understand the ecological factors influencing variability in distributions and to assess potential risk from shipping, fishing, and other human activities. Systematic surveys record species presence and absence, as well as the associated search effort, but are very expensive. Presence-only data consisting only of sightings can increase sample size, but may be biased in both geographical and niche space. We built generalized additive models (GAMs) using presence-absence sightings data and maximum entropy models (Maxent) using the same presence-absence sightings data, and also using presence-only sightings data, for four large whale species in the eastern tropical Pacific Ocean: humpback (Megaptera novaeangliae), blue (Balaenoptera musculus), Bryde's (Balaenoptera edeni), and sperm whales (Physeter macrocephalus). Environmental variables were surface temperature, surface salinity, thermocline depth, stratification index, and seafloor depth. We compared predicted distributions from each of the two model types. Maxent and GAM model predictions based on systematic survey data are very similar, when Maxent absences are selected from the survey trackline data. However, we show that spatial bias in presence-only Maxent predictions can be caused by using pseudo-absences instead of observed absences and by the sampling biases of both opportunistic data and stratified systematic survey data with uneven coverage between strata. Predictions of uncommon large whale distributions from Maxent or other presence-only techniques may be useful for science or management, but only if spatial bias in the observations is addressed in the derivation and interpretation of model predictions.
Idioma originalInglés estadounidense
PublicaciónFrontiers in Marine Science
DOI
EstadoPublicada - 12 nov 2018

Huella dactilar

whales
whale
prediction
modeling
entropy
Entropy
methodology
distribution
comparison
Megaptera novaeangliae
sampling bias
environmental factors
species occurrence
shipping
Freight transportation
thermocline
endangered species
Pacific Ocean
sperm
surface temperature

Citar esto

Fiedler, P. C., Redfern, J. V., Forney, K. A., Palacios, D. M., Sheredy, C., Rasmussen, K., ... Ballance, L. T. (2018). Prediction of large whale distributions: A comparison of presence-absence and presence-only modeling techniques. Frontiers in Marine Science. https://doi.org/10.3389/fmars.2018.00419
Fiedler, Paul C. ; Redfern, Jessica V. ; Forney, Karin A. ; Palacios, Daniel M. ; Sheredy, Corey ; Rasmussen, Kristin ; García-Godos, Ignacio ; Santillán, Luis ; Tetley, Michael J. ; Félix, Fernando ; Ballance, Lisa T. / Prediction of large whale distributions: A comparison of presence-absence and presence-only modeling techniques. En: Frontiers in Marine Science. 2018.
@article{c94d7a0fc56e4f46846077aaa5f1face,
title = "Prediction of large whale distributions: A comparison of presence-absence and presence-only modeling techniques",
abstract = "{\circledC} 2018 Fiedler, Redfern, Forney, Palacios, Sheredy, Rasmussen, Garc{\'i}a-Godos, Santill{\'a}n, Tetley, F{\'e}lix and Ballance. Species distribution models that predict species occurrence or density by quantifying relationships with environmental variables are used for a variety of scientific investigations and management applications. For endangered species, such as large whales, models help to understand the ecological factors influencing variability in distributions and to assess potential risk from shipping, fishing, and other human activities. Systematic surveys record species presence and absence, as well as the associated search effort, but are very expensive. Presence-only data consisting only of sightings can increase sample size, but may be biased in both geographical and niche space. We built generalized additive models (GAMs) using presence-absence sightings data and maximum entropy models (Maxent) using the same presence-absence sightings data, and also using presence-only sightings data, for four large whale species in the eastern tropical Pacific Ocean: humpback (Megaptera novaeangliae), blue (Balaenoptera musculus), Bryde's (Balaenoptera edeni), and sperm whales (Physeter macrocephalus). Environmental variables were surface temperature, surface salinity, thermocline depth, stratification index, and seafloor depth. We compared predicted distributions from each of the two model types. Maxent and GAM model predictions based on systematic survey data are very similar, when Maxent absences are selected from the survey trackline data. However, we show that spatial bias in presence-only Maxent predictions can be caused by using pseudo-absences instead of observed absences and by the sampling biases of both opportunistic data and stratified systematic survey data with uneven coverage between strata. Predictions of uncommon large whale distributions from Maxent or other presence-only techniques may be useful for science or management, but only if spatial bias in the observations is addressed in the derivation and interpretation of model predictions.",
author = "Fiedler, {Paul C.} and Redfern, {Jessica V.} and Forney, {Karin A.} and Palacios, {Daniel M.} and Corey Sheredy and Kristin Rasmussen and Ignacio Garc{\'i}a-Godos and Luis Santill{\'a}n and Tetley, {Michael J.} and Fernando F{\'e}lix and Ballance, {Lisa T.}",
year = "2018",
month = "11",
day = "12",
doi = "10.3389/fmars.2018.00419",
language = "American English",
journal = "Frontiers in Marine Science",
issn = "2296-7745",
publisher = "Frontiers Media",

}

Fiedler, PC, Redfern, JV, Forney, KA, Palacios, DM, Sheredy, C, Rasmussen, K, García-Godos, I, Santillán, L, Tetley, MJ, Félix, F & Ballance, LT 2018, 'Prediction of large whale distributions: A comparison of presence-absence and presence-only modeling techniques', Frontiers in Marine Science. https://doi.org/10.3389/fmars.2018.00419

Prediction of large whale distributions: A comparison of presence-absence and presence-only modeling techniques. / Fiedler, Paul C.; Redfern, Jessica V.; Forney, Karin A.; Palacios, Daniel M.; Sheredy, Corey; Rasmussen, Kristin; García-Godos, Ignacio; Santillán, Luis; Tetley, Michael J.; Félix, Fernando; Ballance, Lisa T.

En: Frontiers in Marine Science, 12.11.2018.

Resultado de la investigación: Contribución a una revistaArtículoInvestigaciónrevisión exhaustiva

TY - JOUR

T1 - Prediction of large whale distributions: A comparison of presence-absence and presence-only modeling techniques

AU - Fiedler, Paul C.

AU - Redfern, Jessica V.

AU - Forney, Karin A.

AU - Palacios, Daniel M.

AU - Sheredy, Corey

AU - Rasmussen, Kristin

AU - García-Godos, Ignacio

AU - Santillán, Luis

AU - Tetley, Michael J.

AU - Félix, Fernando

AU - Ballance, Lisa T.

PY - 2018/11/12

Y1 - 2018/11/12

N2 - © 2018 Fiedler, Redfern, Forney, Palacios, Sheredy, Rasmussen, García-Godos, Santillán, Tetley, Félix and Ballance. Species distribution models that predict species occurrence or density by quantifying relationships with environmental variables are used for a variety of scientific investigations and management applications. For endangered species, such as large whales, models help to understand the ecological factors influencing variability in distributions and to assess potential risk from shipping, fishing, and other human activities. Systematic surveys record species presence and absence, as well as the associated search effort, but are very expensive. Presence-only data consisting only of sightings can increase sample size, but may be biased in both geographical and niche space. We built generalized additive models (GAMs) using presence-absence sightings data and maximum entropy models (Maxent) using the same presence-absence sightings data, and also using presence-only sightings data, for four large whale species in the eastern tropical Pacific Ocean: humpback (Megaptera novaeangliae), blue (Balaenoptera musculus), Bryde's (Balaenoptera edeni), and sperm whales (Physeter macrocephalus). Environmental variables were surface temperature, surface salinity, thermocline depth, stratification index, and seafloor depth. We compared predicted distributions from each of the two model types. Maxent and GAM model predictions based on systematic survey data are very similar, when Maxent absences are selected from the survey trackline data. However, we show that spatial bias in presence-only Maxent predictions can be caused by using pseudo-absences instead of observed absences and by the sampling biases of both opportunistic data and stratified systematic survey data with uneven coverage between strata. Predictions of uncommon large whale distributions from Maxent or other presence-only techniques may be useful for science or management, but only if spatial bias in the observations is addressed in the derivation and interpretation of model predictions.

AB - © 2018 Fiedler, Redfern, Forney, Palacios, Sheredy, Rasmussen, García-Godos, Santillán, Tetley, Félix and Ballance. Species distribution models that predict species occurrence or density by quantifying relationships with environmental variables are used for a variety of scientific investigations and management applications. For endangered species, such as large whales, models help to understand the ecological factors influencing variability in distributions and to assess potential risk from shipping, fishing, and other human activities. Systematic surveys record species presence and absence, as well as the associated search effort, but are very expensive. Presence-only data consisting only of sightings can increase sample size, but may be biased in both geographical and niche space. We built generalized additive models (GAMs) using presence-absence sightings data and maximum entropy models (Maxent) using the same presence-absence sightings data, and also using presence-only sightings data, for four large whale species in the eastern tropical Pacific Ocean: humpback (Megaptera novaeangliae), blue (Balaenoptera musculus), Bryde's (Balaenoptera edeni), and sperm whales (Physeter macrocephalus). Environmental variables were surface temperature, surface salinity, thermocline depth, stratification index, and seafloor depth. We compared predicted distributions from each of the two model types. Maxent and GAM model predictions based on systematic survey data are very similar, when Maxent absences are selected from the survey trackline data. However, we show that spatial bias in presence-only Maxent predictions can be caused by using pseudo-absences instead of observed absences and by the sampling biases of both opportunistic data and stratified systematic survey data with uneven coverage between strata. Predictions of uncommon large whale distributions from Maxent or other presence-only techniques may be useful for science or management, but only if spatial bias in the observations is addressed in the derivation and interpretation of model predictions.

UR - https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85056576002&origin=inward

UR - https://www.scopus.com/inward/citedby.uri?partnerID=HzOxMe3b&scp=85056576002&origin=inward

U2 - 10.3389/fmars.2018.00419

DO - 10.3389/fmars.2018.00419

M3 - Article

JO - Frontiers in Marine Science

JF - Frontiers in Marine Science

SN - 2296-7745

ER -