Spatiotemporal Changes and Influencing Factors of Carbon Storage in the Jinan Metropolitan Area, China, Using the InVEST Model Coupled with XGBoost-SHAP and MGWR Models
InVESTモデルとXGBoost-SHAP、MGWRモデルを組み合わせた中国済南都市圏の炭素貯留の時空間変化と影響要因 (AI 翻訳)
Yubin Liu, Jianfei Cao, Chao Fan, Bing Zhang
🤖 gxceed AI 要約
日本語
本研究は、中国済南都市圏を対象に、2000年から2024年までの土地利用データを用いて、InVESTモデルとXGBoost-SHAP、MGWRモデルを組み合わせ、炭素貯留の時空間変化とその影響要因を分析した。その結果、農地から建設用地への転換が炭素貯留減少の主因であり、自然要因が人為的要因よりも強く影響していることが明らかになった。これらの知見は、都市計画や生態保全に貢献する。
English
This study analyzes spatiotemporal changes in carbon storage in the Jinan metropolitan area from 2000 to 2024 using the InVEST model coupled with XGBoost-SHAP and MGWR models. It finds that cropland-to-construction conversion is the main cause of carbon loss, with natural factors dominating over human activities. The results inform urban planning and ecological conservation.
Unofficial AI-generated summary based on the public title and abstract. Not an official translation.
📝 gxceed 編集解説 — Why this matters
日本のGX文脈において
中国の「双炭」戦略に基づく地域研究であり、日本の都市計画や土地利用政策にも示唆を与える。特に、SSBJやカーボンニュートラルに向けた自治体の取り組みにおいて、土地利用変化と炭素貯留の関係を評価する手法として参考になる。
In the global GX context
This study provides empirical evidence on land-use-driven carbon storage changes, relevant to global climate mitigation and urban sustainability. It demonstrates a methodological framework that can be applied in other metropolitan areas, contributing to the broader discourse on land-use planning and carbon neutrality.
👥 読者別の含意
🔬研究者:土地利用と炭素貯留の関係を分析するためのモデル統合手法を学ぶことができる。
🏢実務担当者:都市計画や環境アセスメントにおいて、炭素貯留を考慮した土地利用計画の策定に活用できる。
🏛政策担当者:地域の炭素貯留を保全するための政策立案に、科学的根拠を提供する。
📄 Abstract（原文）
Within the framework of the dual carbon strategy, investigating the spatiotemporal characteristics and driving factors of carbon sequestration in metropolitan areas through land use analysis is important for mitigating climate change and promoting regional ecological protection and sustainable development. On the basis of land use time points for five phases from the Jinan metropolitan area (JMA) covering the period from 2000 to 2024, the Integrated Valuation of Ecosystem Services and Tradeoffs (InVEST) model was coupled with the extreme gradient boosting (XGBoost)–Shapley Additive exPlanations (SHAP) and multiscale geographically weighted regression (MGWR) models to explore the spatiotemporal variations in carbon storage and its driving factors. In the last 24 years, cropland has been the predominant land use category in the JMA, representing almost 62% of the overall area. Throughout the five periods, the transition from cropland to construction land predominated, resulting in an 11.78% reduction in farmland and a 49.73% expansion in construction land. Between 2000 and 2024, carbon storage in the JMA decreased overall, with a total reduction of 3.70 Tg. The occupation of farmland for construction purposes was the primary cause of the decrease in carbon storage. The spatial pattern of carbon storage was similar to that of land use in the JMA, characterized by a distribution pattern with elevated values in the southeast and reduced values in the northwest. The SHAP analysis results demonstrated that the contributions of driving factors such as elevation, vegetation coverage, human footprint, and population density were generally high, making them the main drivers affecting carbon storage, with a significantly greater contribution of natural factors than human activity factors. The MGWR model results revealed that the digital elevation model and fractional vegetation cover positively influenced carbon storage in the JMA, whereas the population density imposed a negative effect. These results could guide the judicious allocation and utilisation of resources in urban regions, the establishment of ecological conservation areas, and the advancement of regional sustainability.
🔗 Provenance — このレコードを発見したソース
- openalex https://doi.org/10.3390/su18168321first seen 2026-08-16 05:07:05
🔔 こうした論文の新着を逃したくない方は キーワードアラート に登録（無料・3キーワードまで）。
gxceed は公開メタデータに基づく研究支援データセットです。要約・翻訳・解説は AI 支援で生成されています。 最終的な解釈・検証は利用者が原典資料に基づいて行うことを前提とします。