Remote sensing of terrestrial gross primary productivity: a review of advances in theoretical foundation, key parameters and methods

Abstract

Accurately estimating gross primary productivity (GPP), the largest carbon flux in terrestrial ecosystems, is crucial for advancing our understanding of global carbon cycle and predicting climate feedbacks. The advancements in remote sensing (RS) have facilitated the development of GPP estimation models at regional and global scales in recent decades. This article systemically reviews the development of RS-based GPP estimation in three main aspects: theoretical foundation, key parameters and methods. Regarding the theoretical foundation, RS generally excels in representing key characteristics during the light transmission process of photosynthesis. However, it exhibits a relatively weaker ability to describe the carbon reaction process, severely limiting the in-depth understanding of the mechanisms of RS-based GPP estimation. Concerning key parameters, the definition of traditional parameters, such as leaf area index (LAI), photosynthetically active radiation (PAR), and fraction of absorbing PAR, has been detailed in the development of RS (e.g. LAI is divided into sunlit LAI and shaded LAI). However, their accuracy still needs improvement. Additionally, researchers have developed effective parameters (e.g. photochemical reflectance index, sun-induced chlorophyll fluorescence, and the maximum carboxylation rate) that possess increased capability to represent and interpret the carbon reaction process of photosynthesis. Regarding estimation methods, although the four main categories of RS-based GPP estimation models (statistical model, light use efficiency model, RS-based process model and machine learning-based model) have made significant progress in parameter optimization, the estimation accuracy and mechanism innovation remain less than satisfactory. Finally, we summarize the current issues of RS-based GPP estimation related to parameters performance and accuracy, model mechanisms and capabilities, as well as scale and connotation mismatch. Integrating more adequate in situ and comprehensive observations would enhance the interpretability of GPP estimation models, providing more reliable insights into the mechanisms in future studies. This article contributes to understanding of the photosynthetic process and RS-based GPP estimation, potentially aiding in the development of parameter optimization (improving the estimation accuracy of existing parameters and developing new ones) and model design (introducing new parameters and exploring new mechanistic models).

Keywords

• Gross primary production (GPP)
• light use efficiency (LUE)
• photosynthetic process
• remote sensing
• sun-induced chlorophyll fluorescence (SIF)