Date of Award

Spring 5-7-2021

Level of Access Assigned by Author

Open-Access Thesis

Degree Name

Master of Science (MS)


Marine Biology


Yong Chen

Second Committee Member

Gayle Zydlewski

Third Committee Member

Bai Li


The Gulf of Maine (GOM) is a highly complex environment and previous studies have suggested needs to account for spatial nonstationarity in species distribution models for the American lobster (Homarus americanus). Spatial nonstationarity can be defined as the presence of variation in relationships between independent and dependent variables across space (Windle et al., 2012). To explore impacts of spatial nonstationarity on species distribution, models with the following three assumptions were compared: (1) stationary relationships between species distributions and environmental variables; (2) nonstationary density-environment relationships between eastern and western GOM, and (3) nonstationary density-environment relationships across eastern, central, and western GOM. These comparisons were made amongst generalized additive models (GAMs) to evaluate estimations in lobster spatial distribution, and habitat suitability index (HSI) models to evaluate estimations in lobster habitat suitability. The spatial scales used in these models were largely determined by the GOM coastal currents. Lobster data were sourced from the Maine-New Hampshire Inshore Bottom Trawl Survey from years 2000-2019. We considered spatial and environmental variables including latitude and longitude, bottom temperature, bottom salinity, distance from shore, and sediment grain size in this study. The lobster data utilized in this study were divided into eight groups based on season (fall and spring), sex (female and male), and size (juveniles and adults). Estimates of spatial density and habitat suitability distributions were made for the hindcasting years of 2000, 2006, 2012, 2017, and for the forecasting time period 2028- iii 2055 under the Representative Concentration Pathway (RPC) 8.5 “business as usual” climate warming scenario. We found that the model with the finest scale performed best in both model types tested. This suggests that accounting for spatial nonstationarity in the GOM leads to improved spatial distribution and habitat suitability estimates. Forecasted species distribution estimates revealed that stationary models tended to comparatively overestimate (IQR≅ -36 to 0%) most season 𝗑 sex 𝗑 size group lobster abundances in western GOM, underestimate in the western portion of central GOM, and overestimate in the eastern portion of central GOM (IQR≅ -66 to 29%), with slightly less consistent and patchy trends amongst groups in eastern GOM (IQR≅ -15 to 62% for model 1:2 comparisons and IQR≅ -31 to 28% for model 1:3 comparisons). While in forecasted HSI model estimates, stationary models tended to comparatively overestimate the suitability of habitat for juvenile lobsters (IQR≅ -28 to 1%). For adult lobsters, stationary models estimated higher suitability in both coastal waters in western GOM (IQR≅ -7 to 14%) and farther offshore waters in eastern GOM (IQR≅ -2 to 13% for model 1:2 comparison and IQR≅ -6 to 12% for model 1:3 comparison) than nonstationary models applied at finer scales. Stationary adult HSI models also estimated lower suitability in coastal eastern GOM waters and some offshore western GOM waters as well. The estimated results from stationary and nonstationary GAMs and HSI models were statistically different (p