This review summarizes the progress and challenges in remotely sensed terrestrial carbon fluxes. Satellite-based data and methodologies have been synthesized to estimate the main flux components of terrestrial carbon balance over the past two decades. The global gross primary productivity (GPP) from 2001–2022 is 134 ± 14 PgC yr⁻¹, with nearly half occurring in tropical regions. Using satellite-based atmospheric inversion, less than 2% of global GPP is converted into a net carbon sink of 2.28 ± 1.12 PgC yr⁻¹, comparable to stock change-based estimates but twice as large as model-based estimates. By decomposing satellite-derived net carbon balance, it was inferred that ~43% of global GPP is respired through soil microbes, higher than previous bottom-up estimates. Accurate remote sensing of terrestrial carbon balance requires enhancing representations of photosynthetic responses to rising CO₂ and disturbances, developing satellite-constrained below-ground carbon dynamics, and separating natural fluxes from anthropogenic CO₂ emissions. The terrestrial carbon cycle is crucial for understanding land carbon sinks and climate mitigation measures. Terrestrial ecosystem modeling provides spatially and temporally explicit estimates of carbon fluxes. Advances in process-based models and ground-based observations have improved understanding of terrestrial carbon fluxes. The Global Carbon Project's annual carbon budget estimates terrestrial carbon fluxes from Dynamic Global Vegetation Models (DGVMs). However, model-based estimates have large uncertainties due to poor representation of processes and observational constraints. Satellite-based retrieval of terrestrial carbon fluxes involves estimating global primary productivity, anthropogenic and natural disturbances, and net biome production. Remote sensing has been used for GPP estimates through process-based models, light use efficiency models, machine learning models, and empirical relationships with vegetation indicators. The use of satellite observations to estimate terrestrial carbon fluxes provides an independent approach for real-time carbon balance accounting. However, inherent uncertainties, such as saturation issues, can restrict the ability of remote sensing to provide accurate estimates. The current status of satellite-derived terrestrial carbon fluxes shows a general increasing trend in global GPP, with a mean value of 0.45 ± 0.11 PgC yr⁻². The magnitude of the trend varies across different satellite products. Spatial patterns show an almost ubiquitous increase in GPP, except in Australia and Central Asia. The inter-product discrepancy in GPP trends becomes larger in South America and Africa. The net effect of historical land use and land use change was a carbon source of 1.2 ± 0.7 PgC yr⁻¹ for the most recent decade. Land-use change emissions mainly come from tropical Africa, Latin America, and South and Southeast Asia, while carbon uptakes are found in North America, Europe, former Soviet Union countries, and China. Wildland fires are a widespread disturbance, with remote sensing-based burned area providing fundamental data for quantifying global and regional fire carbonThis review summarizes the progress and challenges in remotely sensed terrestrial carbon fluxes. Satellite-based data and methodologies have been synthesized to estimate the main flux components of terrestrial carbon balance over the past two decades. The global gross primary productivity (GPP) from 2001–2022 is 134 ± 14 PgC yr⁻¹, with nearly half occurring in tropical regions. Using satellite-based atmospheric inversion, less than 2% of global GPP is converted into a net carbon sink of 2.28 ± 1.12 PgC yr⁻¹, comparable to stock change-based estimates but twice as large as model-based estimates. By decomposing satellite-derived net carbon balance, it was inferred that ~43% of global GPP is respired through soil microbes, higher than previous bottom-up estimates. Accurate remote sensing of terrestrial carbon balance requires enhancing representations of photosynthetic responses to rising CO₂ and disturbances, developing satellite-constrained below-ground carbon dynamics, and separating natural fluxes from anthropogenic CO₂ emissions. The terrestrial carbon cycle is crucial for understanding land carbon sinks and climate mitigation measures. Terrestrial ecosystem modeling provides spatially and temporally explicit estimates of carbon fluxes. Advances in process-based models and ground-based observations have improved understanding of terrestrial carbon fluxes. The Global Carbon Project's annual carbon budget estimates terrestrial carbon fluxes from Dynamic Global Vegetation Models (DGVMs). However, model-based estimates have large uncertainties due to poor representation of processes and observational constraints. Satellite-based retrieval of terrestrial carbon fluxes involves estimating global primary productivity, anthropogenic and natural disturbances, and net biome production. Remote sensing has been used for GPP estimates through process-based models, light use efficiency models, machine learning models, and empirical relationships with vegetation indicators. The use of satellite observations to estimate terrestrial carbon fluxes provides an independent approach for real-time carbon balance accounting. However, inherent uncertainties, such as saturation issues, can restrict the ability of remote sensing to provide accurate estimates. The current status of satellite-derived terrestrial carbon fluxes shows a general increasing trend in global GPP, with a mean value of 0.45 ± 0.11 PgC yr⁻². The magnitude of the trend varies across different satellite products. Spatial patterns show an almost ubiquitous increase in GPP, except in Australia and Central Asia. The inter-product discrepancy in GPP trends becomes larger in South America and Africa. The net effect of historical land use and land use change was a carbon source of 1.2 ± 0.7 PgC yr⁻¹ for the most recent decade. Land-use change emissions mainly come from tropical Africa, Latin America, and South and Southeast Asia, while carbon uptakes are found in North America, Europe, former Soviet Union countries, and China. Wildland fires are a widespread disturbance, with remote sensing-based burned area providing fundamental data for quantifying global and regional fire carbon