TY - GEN
T1 - Geostatistical Sampling Designs Under Preferential Sampling for Black Scabbardfish
AU - Simões, Paula
AU - Carvalho, Maria Lucília
AU - Figueiredo, Ivone
AU - Monteiro, Andreia
AU - Natário, Isabel
N1 - Funding Information:
info:eu-repo/grantAgreement/FCT/3599-PPCDT/PTDC%2FMAT-STA%2F28243%2F2017/PT#
info:eu-repo/grantAgreement/FCT/6817 - DCRRNI ID/UIDB%2F00297%2F2020/PT#
info:eu-repo/grantAgreement/FCT/6817 - DCRRNI ID/UIDP%2F00297%2F2020/PT#
info:eu-repo/grantAgreement/FCT/6817 - DCRRNI ID/UIDB%2F00006%2F2020/PT#
info:eu-repo/grantAgreement/FCT/6817 - DCRRNI ID/UIDP%2F00006%2F2020/PT#
Publisher Copyright:
© 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.
PY - 2022/11/29
Y1 - 2022/11/29
N2 - In Portugal, the spatial distribution and abundance of the black scabbardfish (BSF) is mostly unknown, the existing information relying on data from commercial fisheries. Available data refers to areas where fisherman expect to have higher catches of the species, resulting in fishing locations that are not selected randomly but preferentially. The BSF captures in Portuguese waters were previously modelled, taking the sampling preferentiality into account, using a Bayesian approach and INLA methodology, considering stochastic partial differential equations (SPDE) for geostatistical data, jointly with a Log-Cox point process model. Based on this work, the aim of this study is to construct a new survey design to improve the BSF capture estimates and to analyse the effect of preferential sampling on the choice of new sampling locations and its influence in the sampling design choice. Within this approach, different design classes are investigated, namely random, simple inhibitory and adaptive geostatistical sampling designs, regarding the problem of spatial prediction, in order to achieve the optimal BSF design towards the objective of the analysis.
AB - In Portugal, the spatial distribution and abundance of the black scabbardfish (BSF) is mostly unknown, the existing information relying on data from commercial fisheries. Available data refers to areas where fisherman expect to have higher catches of the species, resulting in fishing locations that are not selected randomly but preferentially. The BSF captures in Portuguese waters were previously modelled, taking the sampling preferentiality into account, using a Bayesian approach and INLA methodology, considering stochastic partial differential equations (SPDE) for geostatistical data, jointly with a Log-Cox point process model. Based on this work, the aim of this study is to construct a new survey design to improve the BSF capture estimates and to analyse the effect of preferential sampling on the choice of new sampling locations and its influence in the sampling design choice. Within this approach, different design classes are investigated, namely random, simple inhibitory and adaptive geostatistical sampling designs, regarding the problem of spatial prediction, in order to achieve the optimal BSF design towards the objective of the analysis.
KW - Geostatistics
KW - INLA
KW - Preferential sampling
KW - Sampling design
UR - http://www.scopus.com/inward/record.url?scp=85144357105&partnerID=8YFLogxK
U2 - 10.1007/978-3-031-12766-3_11
DO - 10.1007/978-3-031-12766-3_11
M3 - Conference contribution
AN - SCOPUS:85144357105
SN - 978-3-031-12765-6
T3 - Springer Proceedings in Mathematics and Statistics
SP - 137
EP - 151
BT - Recent Developments in Statistics and Data Science
A2 - Bispo, Regina
A2 - Henriques-Rodrigues, Lígia
A2 - Alpizar-Jara, Russell
A2 - de Carvalho, Miguel
PB - Springer
CY - Cham
T2 - 25th Congress of the Portuguese Statistical Society, SPE 2021
Y2 - 13 October 2021 through 16 October 2021
ER -