Data integration improves species distribution forecasts under novel ocean conditions

Accurate forecasts of species distributions in response to a changing climate are essential for proactive management and conservation decision-making. However, species distribution models (SDMs) often have limited capacity to produce robust forecasts under novel environmental conditions, partly due to limitations in model training data. Model-based approaches that leverage diverse types of data have advanced over the last decade, yet their forecasting skill, especially during episodic climatic events, remains uncertain. Here, we develop a suite of SDMs for a commercially important fishery species, albacore tuna (Thunnus alalunga), to evaluate forecast skill under marine heatwave conditions. We compare models that use different methods to leverage data sources (data pooling vs. joint likelihood) and to address spatial dependence (environmental and spatial effects vs. environmental-only) to assess their relative performance in predicting species distributions under novel environmental con..., , # Data integration improves species distribution forecasts under novel ocean conditions

This repository contains model outputs to help recreate visualizations and interpret results. Specifically, the repo contains:

1. Model performance outputs (i.e., AUC, MAE, Boyce, & Novelty metrics) from retrospective monthly forecast analysis
2. Each model's spatial forecast prediction for an example month (September 2015)
3. Response curves (i.e., marginal effects) from each model

**Note:** The data provided in this repository do not contain the raw data, but only model-derived outputs, due to restrictions associated with both datasets used in the study.

Specifically, logbook data for the U.S. albacore troll and pole-and-line fishery are confidential U.S. government data and are not publicly available. The raw data cannot be made public under the Magnuson–Stevens Fishery Conservation and Management Reauthorization Act of 2006, section 402(b), 16 U.S.C. 1881a. To request access to U.S. Highly Mi...,

精准预测气候变化下的物种分布，对于开展主动式管理与保护决策至关重要。然而，物种分布模型（species distribution models, SDMs）在全新环境条件下往往难以生成可靠的预测结果，这在一定程度上源于模型训练数据的局限性。近十年来，依托多源数据的模型方法取得了长足进展，但它们的预测能力——尤其是在极端气候事件期间——仍未明确。本研究针对一种具有重要商业价值的渔业物种长鳍金枪鱼（Thunnus alalunga）构建了一系列SDMs，以评估其在海洋热浪条件下的预测能力。我们对比了采用不同数据利用策略（数据合并法 vs. 联合似然法）以及空间相关性处理方式（环境与空间效应法 vs. 仅环境效应法）的模型，以评估它们在全新环境条件下预测物种分布的相对性能。

# 数据集成提升全新海洋条件下的物种分布预测能力

本仓库包含可用于复现可视化成果与解读研究结果的模型输出文件，具体包括：
1. 基于回顾性月度预测分析得到的模型性能指标（即AUC、MAE、博伊斯指数Boyce与新颖性指标）；
2. 各模型在示例月份（2015年9月）的空间分布预测结果；
3. 各模型的响应曲线（即边际效应）。

**注意：** 由于本研究使用的两类数据集均存在使用限制，本仓库提供的数据仅包含模型衍生输出，不包含原始数据。
具体而言，美国长鳍金枪鱼拖钓和竿钓渔业的捕捞日志数据属于美国政府机密数据，无法公开获取。根据《2006年马格努森-史蒂文斯渔业保护和管理重新授权法案》第402(b)条（16 U.S.C. 1881a）的规定，原始数据不得公开。如需申请获取美国高度混合……