Deep Learning with Pretrained Framework Unleashes the Power of Satellite-Based Global Fine-Mode Aerosol Retrieval.

Fine-mode aerosol optical depth (fAOD) is a vital proxy for the concentration of anthropogenic aerosols in the atmosphere. Currently, the limited data length and high uncertainty of the satellite-based data diminish the applicability of fAOD for climate research. Here, we propose a novel pretrained deep learning framework that can extract information underlying each satellite pixel and use it to create new latent features that can be employed for improving retrieval accuracy in regions without in situ data. With the proposed model, we developed a new global fAOD (at 0.5 µm) data from 2001 to 2020, resulting in a 10% improvement in the overall correlation coefficient (R) during site-based independent validation and a 15% enhancement in non-AERONET site areas validation. Over the past two decades, there has been a noticeable downward trend in global fAOD (-1.39 × 10-3/year). Compared to the general deep-learning model, our method reduces the global trend's previously overestimated magnitude by 7% per year. China has experienced the most significant decline (-5.07 × 10-3/year), which is 3 times greater than the global trend. Conversely, India has shown a significant increase (7.86 × 10-4/year). This study bridges the gap between sparse in situ observations and abundant satellite measurements, thereby improving predictive models for global patterns of fAOD and other climate factors.

Vegetation increases global climate vulnerability risk by shifting climate zones in response to rising atmospheric CO2.

Global climate zones are experiencing widespread shifts with ongoing rise in atmospheric CO2, influencing vegetation growth and shifting its distributions to challenge ecosystem structure and function, posing threats on ecological and societal safety. However, how rising atmospheric CO2 affects the pace of global climate zone shifts is highly uncertain. More attentions are urgently required to understand the underlying mechanisms and quantifications of regional climate vulnerability in response to rising CO2. In this study, we employ nine Earth system models from CMIP6 to investigate global climate zone shifts with rising CO2, unravel the effects of vegetation physiological response (PHY), and categorize climate vulnerable regions depending on the extent of climate zone shifts. We find that climate zone shifts over half of the global land area, 16.8% of which is contributed by PHY at 4 × CO2. Intriguingly, besides warming, PHY-induced precipitation changes and their interactions with warming dominate about two-fifths of PHY-forced shifts, providing potential direction for model improvement in future predictions of climate zone shifts. Aided with PHY effects, 4 × CO2 imposes substantial climate zone shifts over about one-fifth of the global land area, suggesting substantial changes in local climate and ecosystem structure and functions. Hence, those regions would experience strong climate vulnerability, and face high risk of climate extremes, water scarcity and food production. Our results quantitatively identify the vulnerable regions and unravel the underlying drivers, providing scientific insights to prioritize conservation and restoration efforts to ensure ecological and social safety globally.

Reversed asymmetric warming of sub-diurnal temperature over land during recent decades.

In the latter half of the twentieth century, a significant climate phenomenon "diurnal asymmetric warming" emerged, wherein global land surface temperatures increased more rapidly during the night than during the day. However, recent episodes of global brightening and regional droughts and heatwaves have brought notable alterations to this asymmetric warming trend. Here, we re-evaluate sub-diurnal temperature patterns, revealing a substantial increase in the warming rates of daily maximum temperatures (Tmax), while daily minimum temperatures have remained relatively stable. This shift has resulted in a reversal of the diurnal warming trend, expanding the diurnal temperature range over recent decades. The intensified Tmax warming is attributed to a widespread reduction in cloud cover, which has led to increased solar irradiance at the surface. Our findings underscore the urgent need for enhanced scrutiny of recent temperature trends and their implications for the wider earth system.

Disentangling the effects of vapor pressure deficit on northern terrestrial vegetation productivity.

The impact of atmospheric vapor pressure deficit (VPD) on plant photosynthesis has long been acknowledged, but large interactions with air temperature (T) and soil moisture (SM) still hinder a complete understanding of the influence of VPD on vegetation production across various climate zones. Here, we found a diverging response of productivity to VPD in the Northern Hemisphere by excluding interactive effects of VPD with T and SM. The interactions between VPD and T/SM not only offset the potential positive impact of warming on vegetation productivity but also amplifies the negative effect of soil drying. Notably, for high-latitude ecosystems, there occurs a pronounced shift in vegetation productivity's response to VPD during the growing season when VPD surpasses a threshold of 3.5 to 4.0 hectopascals. These results yield previously unknown insights into the role of VPD in terrestrial ecosystems and enhance our comprehension of the terrestrial carbon cycle's response to global warming.

Cooperative simultaneous inversion of satellite-based real-time PM2.5 and ozone levels using an improved deep learning model with attention mechanism.

Ground-level fine particulate matter (PM2.5) and ozone (O3) are air pollutants that can pose severe health risks. Surface PM2.5 and O3 concentrations can be monitored from satellites, but most retrieval methods retrieve PM2.5 or O3 separately and disregard the shared information between the two air pollutants, for example due to common emission sources. Using surface observations across China spanning 2014-2021, we found a strong relationship between PM2.5 and O3 with distinct spatiotemporal characteristics. Thus, in this study, we propose a new deep learning model called the Simultaneous Ozone and PM2.5 inversion deep neural Network (SOPiNet), which allows for daily real-time monitoring and full coverage of PM2.5 and O3 simultaneously at a spatial resolution of 5 km. SOPiNet employs the multi-head attention mechanism to better capture the temporal variations in PM2.5 and O3 based on previous days' conditions. Applying SOPiNet to MODIS data over China in 2022, using 2019-2021 to construct the network, we found that simultaneous retrievals of PM2.5 and O3 improved the performance compared with retrieving them independently: the temporal R2 increased from 0.66 to 0.72 for PM2.5, and from 0.79 to 0.82 for O3. The results suggest that near-real time satellite-based air quality monitoring can be improved by simultaneous retrieval of different but related pollutants. The codes of SOPiNet and its user guide are freely available online at https://github.com/RegiusQuant/ESIDLM.

Worldwide impacts of atmospheric vapor pressure deficit on the interannual variability of terrestrial carbon sinks.

Interannual variability of the terrestrial ecosystem carbon sink is substantially regulated by various environmental variables and highly dominates the interannual variation of atmospheric carbon dioxide (CO2) concentrations. Thus, it is necessary to determine dominating factors affecting the interannual variability of the carbon sink to improve our capability of predicting future terrestrial carbon sinks. Using global datasets derived from machine-learning methods and process-based ecosystem models, this study reveals that the interannual variability of the atmospheric vapor pressure deficit (VPD) was significantly negatively correlated with net ecosystem production (NEP) and substantially impacted the interannual variability of the atmospheric CO2 growth rate (CGR). Further analyses found widespread constraints of VPD interannual variability on terrestrial gross primary production (GPP), causing VPD to impact NEP and CGR. Partial correlation analysis confirms the persistent and widespread impacts of VPD on terrestrial carbon sinks compared to other environmental variables. Current Earth system models underestimate the interannual variability in VPD and its impacts on GPP and NEP. Our results highlight the importance of VPD for terrestrial carbon sinks in assessing ecosystems' responses to future climate conditions.

Arctic amplification modulated by Atlantic Multidecadal Oscillation and greenhouse forcing on multidecadal to century scales.

Enhanced warming in the Arctic (Arctic amplification, AA) in the last decades has been linked to several factors including sea ice and the Atlantic Multidecadal Oscillation (AMO). However, how these factors contributed to AA variations in a long-term perspective remains unclear. By reconstructing a millennial AA index combining climate model simulations with recently available proxy data, this work determines the important influences of the AMO and anthropogenic greenhouse gas forcing on AA variations in the last millennium, leading to identification of a significant downward trend of AA on top of a sustained strong AMO modulation at the multidecadal scales. The decreased AA during the industrial era was strongly associated with the anthropogenic forcing, proving the emerging role of the forcing in reducing the AA strength.