Diverse biosphere influence on carbon and heat in mixed urban Mediterranean landscape revealed by high resolution thermal and optical remote sensing

A fundamental challenge in verifying urban CO2 emissions reductions is estimating the biological influence that can confound emission source attribution across heterogeneous and diverse landscapes. Recent work using atmospheric radiocarbon revealed a substantial seasonal influence of the managed urban biosphere on regional carbon budgets in the Los Angeles megacity, but lacked spatially explicit attribution of the diverse biological influences needed for flux quantification and decision making. New optical and thermal sensors from airborne and satellite platforms simultaneously resolve fine-scale urban land cover, land use (irrigation), phenology and water stress effects required to accurately quantify biological influences on CO2 exchange in complex urban environments. We use remote sensing constraints with an urban land surface model to quantify spatial and seasonal gross primary production (GPP) variability across the Southern California Air Basin (SoCAB) at 30 m resolution. High spatial resolution optical and thermal remote sensing imagery provides attribution of landscape influences related to vegetation type, fragmentation and irrigation across urban and non-urban gradients. Results showed that montane non-urban shrubs and trees dominate the regional carbon budget of SoCAB (80% of GPP). The urban carbon budget is strongly influenced by land use and landscape fragmentation. Irrigated vegetation accounts for only 21% of urban vegetation but 31% of annual GPP, driven by turf grass, and is twice as productive as non-irrigated vegetation during hot and dry Mediterranean summer months. Fragmented vegetation accounts for 25% of urban vegetation but 50% of annual GPP, with cooler interior vegetation enhancing tree and grass GPP in spring and summer, and irrigation mitigating stress-driven GPP loss in edge vegetation. Our satellite driven, very high resolution modeling framework highlights the important influence of land use and fragmentation on urban carbon budgets, and provides a path forward for quantifying urban biospheric influences globally.验证城市二氧化碳（CO2）减排成效的一项核心挑战，在于估算异质多样景观中会干扰排放源归因的生物影响。此前有研究利用大气放射性碳（radiocarbon）发现，管理型城市生物圈对洛杉矶大都市区的区域碳收支存在显著的季节影响，但该研究未能针对通量量化与决策制定所需的多样生物影响，实现空间显性的归因分析。本次研究采用的机载与卫星平台搭载的新型光学与热红外传感器（optical and thermal sensors），可同步解析精细尺度的城市土地覆盖、土地利用（land use）与灌溉（irrigation）、物候（phenology）以及水分胁迫效应，这些要素正是精准量化复杂城市环境中二氧化碳交换相关生物影响的必要条件。本研究结合遥感约束与城市地表模型，以30米分辨率量化了南加州空气盆地（Southern California Air Basin，SoCAB）内总初级生产力（GPP）的空间与季节变异特征。高空间分辨率的光学与热红外遥感影像，可实现城市与非城市梯度下与植被类型、景观破碎化及灌溉相关的景观影响归因分析。研究结果显示，山地非城市灌丛与树木覆盖主导了南加州空气盆地的区域碳收支（占总初级生产力的80%）。城市碳收支则显著受土地利用与景观破碎化的影响。灌溉型植被仅占城市植被总量的21%，却贡献了年度总初级生产力的31%，这一现象由草坪植被驱动；在地中海气候炎热干燥的夏季，灌溉型植被的生产力是非灌溉型植被的两倍。破碎化植被占城市植被总量的25%，却贡献了年度总初级生产力的50%；春季与夏季时，植被内部较凉爽的区域可提升树木与草本植被的总初级生产力，而灌溉则可缓解边缘植被因胁迫导致的总初级生产力损失。本研究提出的卫星驱动超高分辨率建模框架，凸显了土地利用与景观破碎化对城市碳收支的重要影响，也为全球范围内城市生物圈影响的量化研究提供了可行路径。