気候リスクが林業の生態系全要素生産性に与える影響：ダブルマシンラーニングによる実証
The impact of climate risk on forestry ecological total factor productivity: evidence from double machine learning (原題)
Liu-Qing Wang, Hang Chen, Wei-Han Zhu, Yan-Jin Zhou
🤖 gxceed AI 要約
日本語
中国30省の2012〜2023年パネルデータを用い、極端気象を中心とする物理的気候リスク指数（CPRI）を構築し、ダブルマシンラーニング（DML）で林業の生態系全要素生産性（TFP）への影響を推定した。気候リスクは林業TFPを有意に低下させ、森林病害虫・鼠害の防除格差の拡大と一般汚染問題への公的関心の押し出しが媒介経路となる。エネルギー集約産業への資本市場エクスポージャーと地域の炭素排出強度が調整効果を持ち、閾値効果と地域異質性も確認された。適応政策として病害虫防除の強化と極端気象時の環境情報発信の継続が示唆される。
English
Using 2012–2023 panel data for 30 Chinese provinces, the authors build a composite physical climate risk index (CPRI) of extreme cold, heat, rainfall and drought, and apply double/debiased machine learning to estimate effects on forestry ecological total factor productivity (TFP). Climate risk significantly lowers forestry TFP, with negative indirect pathways through wider gaps in forest pest/rodent control and crowding-out of public attention to pollution. Capital-market exposure to energy-intensive industries and regional carbon intensity moderate the association, with threshold and regional heterogeneity. Adaptation policy should strengthen pest monitoring and sustain environmental information flows during extreme events.
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
日本では林業・農林業の気候適応は適応計画やJ-クレジット森林吸収源の文脈で重要性が増すが、本論文は中国省レベルの実証であり、SSBJ・有報開示への直接の示唆は限定的。ただし物理的気候リスクの定量化手法（CPRI＋DML）は、日本企業がTCFD/SSBJで物理リスクを評価する際の方法論的参考になりうる。
In the global GX context
The paper sits at the intersection of physical climate risk quantification and machine learning, offering a methodological template (composite physical risk index + DML) that is relevant to TCFD/ISSB physical-risk assessment and adaptation disclosure. Its focus on forestry TFP and Chinese provinces means it adds empirical evidence from a non-EU/US context, complementing global work on climate-risk transmission to productivity and supply chains.
👥 読者別の含意
🔬研究者:物理的気候リスク指数とDMLを組み合わせた因果推定の設計は、気候リスクが生産性に及ぼす経路分析の方法論的参照になる。
🏢実務担当者:林業・農林サプライチェーンに関わる企業は、病害虫防除と極端気象時の情報発信を適応策として検討する材料になる。
🏛政策担当者:適応政策において病害虫モニタリング体制と地域の構造的エクスポージャーに応じたきめ細かなガバナンスの必要性を示唆する。
📄 Abstract（原文）
Extreme climate events associated with climate change pose a growing threat to forestry production efficiency, yet systematic empirical evidence on their effects on ecological total factor productivity in forestry (forestry TFP) and the underlying transmission pathways remains limited. Using panel data for 30 Chinese provinces from 2012 to 2023, we measure provincial physical climate risk, dominated by acute extreme events, with a composite physical climate risk index (CPRI) comprising exposure to extreme cold, extreme heat, extreme rainfall and extreme drought. The index excludes transition risks arising from policy, technological and market changes during the low-carbon transition. We apply a double/debiased machine learning (DML) model to estimate the effect of climate risk on forestry TFP and conduct exploratory tests of potential transmission pathways, moderating effects and nonlinearities. The results show that climate risk is significantly associated with lower forestry TFP. Negative indirect associations are transmitted through a wider control gap for forest diseases, insect pests and rodents and the crowding-out of public search attention to general pollution issues. Capital-market exposure to energy-intensive industries and regional carbon-emission intensity significantly alter the marginal association with climate risk, which also exhibits threshold effects and regional heterogeneity. These findings indicate that forestry climate-adaptation policy should strengthen monitoring and control of forest diseases, insect pests and rodents and maintain continuous environmental-information communication during extreme climate events, while tailoring governance to regional structural exposure and adaptive capacity.
🔗 Provenance — このレコードを発見したソース
- semanticscholar https://www.frontiersin.org/journals/forests-and-global-change/articles/10.3389/ffgc.2026.1930608/pdffirst seen 2026-10-02 05:38:53
🔔 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。