Carbon stocks across Brazilian macro-ecosystems: ecological structure, methodological uncertainty, and implications for carbon accounting

Tropical and subtropical ecosystems are major carbon reservoirs, yet robust carbon accounting across those landscapes remains limited by environmental heterogeneity, uneven representation of pools, and inconsistent measurement methods. Brazil provides a globally relevant model system for tropical carbon accounting due to its diversity of macro-ecosystems and intense land-use transitions. We synthesized 1,299 standardized records from 74 peer-reviewed studies across six Brazilian macro-ecosystems to examine how carbon pools are distributed among ecosystem types, how these patterns compare with global expectations, how much reported variation reflects ecological versus methodological drivers, and which data gaps most constrain reliable carbon accounting. We integrated records for aboveground biomass, belowground biomass, soil organic carbon, and ecosystem-level carbon, standardized all values to Mg C ha⁻¹, and evaluated variability using hierarchical mixed-effects models with climatic moderators. Carbon storage followed a clear ecological hierarchy: humid forests had the highest aboveground biomass, coastal wetlands had the greatest soil organic carbon, and ecotones had the highest ecosystem-level carbon. Relative to agricultural baselines, peak native-system values were approximately five- to sixteen-fold higher across major pools. The best-performing model included climate moderators and explained most of the observed variation (conditional R² = 0.972), although study-level clustering remained high (intraclass correlation coefficient = 0.965), indicating substantial methodological influence on reported estimates. Sampling effort was also strongly unbalanced: soil organic carbon was the most frequently reported pool (n = 581), whereas belowground biomass remained severely underrepresented (n = 45). These findings provide quantitative baselines for Brazilian macro-ecosystems and support the adoption of standardized, multi-pool, depth-explicit protocols to improve monitoring, reporting, and verification of carbon stocks. Methodological heterogeneity remains a dominant source of variation in tropical carbon‑stock estimates, reinforcing the need for consistent approaches to support credible Nature‑based Climate Solutions (NbCS), robust monitoring, reporting and verification (MRV) systems, and carbon‑based mitigation strategies.热带与亚热带生态系统是主要的碳储库（carbon reservoir），但此类景观的精准碳核算（carbon accounting）仍受限于环境异质性、碳库（carbon pool）代表性不均以及测量方法不统一等问题。巴西拥有多样的大型生态系统（macro-ecosystem）且经历着剧烈的土地利用转型，因此成为全球热带碳核算研究的典型模式系统。本研究整合了巴西6个大型生态系统中74项同行评议研究的1299条标准化记录，旨在探究碳库在不同生态系统类型间的分布格局、该格局与全球预期的对比情况、已报道的变异中有多少源于生态驱动与方法学驱动，以及哪些数据缺口最制约可靠的碳核算工作。我们整合了地上生物量（aboveground biomass）、地下生物量（belowground biomass）、土壤有机碳（soil organic carbon）及生态系统尺度碳储量的相关记录，将所有数值统一标准化为兆克碳每公顷（Mg C ha⁻¹），并采用纳入气候调节因子（climatic moderator）的层级混合效应模型（hierarchical mixed-effects model）对变异程度进行评估。碳储量呈现清晰的生态层级特征：湿润森林（humid forest）的地上生物量最高，沿海湿地（coastal wetland）的土壤有机碳储量最大，而生态交错带（ecotone）的生态系统尺度碳储量最高。相较于农业用地基线（agricultural baseline），主要碳库中原生生态系统（native ecosystem）的峰值储量约为农业用地的5至16倍。表现最优的模型纳入了气候调节因子，能够解释绝大多数观测到的变异（条件决定系数R²=0.972），尽管研究层面的聚类效应仍较高（组内相关系数ICC=0.965），这表明方法学因素对已报道的估算结果存在显著影响。采样强度（sampling effort）也存在严重不均衡：土壤有机碳是报道最多的碳库（n=581），而地下生物量的记录量仍极度匮乏（n=45）。上述研究结果为巴西大型生态系统提供了定量基线，并支持采用标准化、多碳库、明确土层深度的监测方案，以提升碳储量的监测、报告与核查（Monitoring, Reporting and Verification, MRV）工作质量。方法学异质性仍是热带碳储量估算中变异的主要来源，这进一步凸显了采用统一方法的必要性，以支撑可信的基于自然的气候解决方案（Nature-based Climate Solutions, NbCS）、完善的监测报告与核查体系以及基于碳的减排策略。