Historical spatiotemporal trends in global mangrove productivity and its response to the environment : perspectives from multiple satellite-based productivity proxies

Abstract

Mangroves, recognized as highly productive ecosystems, play a crucial role in global carbon cycle despite covering only 2 % of coastal ocean area. Understanding the historical trends of mangrove productivity and its responses to the external environment is crucial. However, the global and regional trends in mangrove productivity and the applicability of productivity proxies in mangrove ecosystems remained unclear. Leveraging eight productivity proxy datasets, including four Gross Primary Productivity (GPP) products, two Solar-Induced Fluorescence (SIF) products, the Near-Infrared Reflectance of vegetation (NIRv), and Leaf Area Index (LAI), this study conducted a comprehensive evaluation on spatiotemporal trends of global mangrove productivity. Through comparison with flux tower observations at three temporal scales (yearly, monthly, and 8-day), the Global OCO-2 SIF (GOSIF), global spatially contiguous SIF (CSIF) and GOSIF-based GPP (GOGPP) products outperformed other productivity proxies in indicating mangrove ecosystem productivity, showing stronger consistency (R²: 0.37–0.70), while NIRv and LAI performed worse (R²: 0.07–0.50). Despite uncertainties in the datasets, collective evidence from all eight products revealed an overall increasing trend in global mangrove productivity since 1982, with a decelerated pace post-2000, reducing to roughly 50 % of the earlier rate. Simulation-based sensitivity and contribution analyses indicated that air temperature (AirT), sea surface temperature (SST), and atmospheric CO<inf>2</inf> were primary contributors to the marked increase in global mangrove productivity, while rising vapor pressure deficit (VPD) and sea surface salinity (SSS) were significant factors leading to a decrease, albeit offsetting only a small portion of the increase. This study provided constructive insights for developing mangrove productivity models and hold significant implications for coastal ecosystem conservation.