Predicting Abundance and Distribution of Narwhal and Harp Seal in Northeast Greenland Using Density Surface Modelling and Boost

Specialeforsvar: Yixin Ma

Titel: Predicting Abundance and Distribution of Narwhal and Harp Seal in Northeast Greenland Using Density Surface Modelling and Boosted Regression Trees

Abstract: Understanding the abundance and distribution patterns of marine mammal species within complex ecosystems is crucial for effective conservation and management strategies. In this study, we investigated the abundance and distribution of narwhals (Monodon monoceros) and harp seals (Pagophilus groenlandicus) in Northea on Density Surface Modeling (DSM) and Boosted Regression Trees (BRT), we navigated the challenges posed by limited data availability and species-specific characteristics.
Our findings revealed distinct distribution patterns for narwhals and harp seals. For narwhals, our integrated modeling approach yielded a precise abundance estimate of 1801 (95% CI: [1487, 2182]) in Dove Bay, a total of 4476 (95% CI: [2706, 8050]) narwhals summering in Northeast Greenland, aligning well with previous research. The distribution of narwhals exhibited sensitivity to
factors such as food resources, environmental conditions, and ice dynamics, shedding light on their adaptive behavior in the Northeast Greenland marine environment.
Harp seal distribution, addressed through a transition from DSM to Generalized Additive Model (GAM), indicated preferences for specific habitats, including Southern Dove Bay and around Jøkel Bay. Bathymetry, temperature and prey availability emerged as pivotal drivers shaping their summering patterns.
Notably, contrasting abundance estimates between GAM result, 88365 (95% CI: [68004, 257704]) and BRT result, 136686 (95% CI: [135658, 137486]) underscored the influence of modeling methods on predictions.

Comparing our modeling strategies highlighted the trade-offs between DSM and BRT. We found that DSM offers predictive power and flexibility and BRT excels in predictive accuracy, each approach possesses unique advantages and considerations.
Our study advances the understanding of marine mammal populations in Northeastern Greenland. By unraveling the complex dynamics governing their distribution, our findings contribute to the broader field of marine ecology and inform strategies to safeguard these remarkable species in a rapidly changing marine ecosystem.

Vejledere: Susanne Ditlevsen,
Rikke Guldborg Hansen, Greenland Institute of Natural Resources
Censor: Uffe Høgsbro Thygesen, DTU