Hail is richly recorded in Chinese historical meteorological literature, reflecting the complex interplay between climate change and human society. This study reveals an obvious pattern of exponential growth in hail records (from 886 BCE to 1949 CE), spanning over 2800 yr from the Zhou Dynasty and featuring several peaks within this trend. Each of these peaks lasted approximately 50–150 yr, with intervals between two peaks ranging roughly from 50 to 150 yr. The number of hail records was relatively high during the dynasty of governance period such as the Han, Tang, Northern Song, and Ming. When the country was unified. Although population number shows the strongest positive correlation with hail records, a more intriguing finding is that the epidemic frequency and temperature anomaly have proven to be significant positive and negative influencing factors on hail records. Notably, medium-size epidemic frequency shows a stronger correlation with hail records than do total, large, and extra-large size epidemics. Other factors, including humidity conditions and fiscal balance, also exert some influence on hail records, albeit to a lesser extent. These findings reveal that population number, driven by climate impacts on agriculture and socio-political stability, interacts with temperature anomaly influencing hail formation through atmospheric conditions and epidemic frequency, which is linked through the hail-epidemic disaster chain, collectively shaping historical hail records through interconnected natural and human processes.
Despite the critical role of lakes in the global carbon cycle, simultaneous measurements of δ13C in both dissolved CO2 and dissolved inorganic carbon (DIC) remain scarce, thereby hindering accurate interpretation of δ13C in lake inorganic carbon cycle studies. In this study, we measured δ13C of DIC in water samples across different lakes of China, supplemented by published δ13C data of CO2 that were collected and measured simultaneously. A strong positive correlation (R2=0.75, p<0.01) is observed between δ13C-CO2 and δ13C-DIC, but δ13C-CO2 are generally more negative (avg. −3.6‰, ranging from −8.8‰ to −0.1‰) than δ13C-DIC. While δ13C variations in CO2/DIC are affected by factors such as pH, temperature, DIC concentration, and precipitation (with a stronger effect in arid than in humid regions), a negative correlation between CO2 concentration and δ13C-DIC demonstrates that carbon isotopes can serve as a potential proxy for tracing CO2 variations at a continental scale. Within individual regions, δ13C of CO2/DIC are relatively clustered, indicating that δ13C of inorganic carbon in lake water is primarily shaped by specific regional backgrounds (e.g., external carbon inputs or local hydrochemical conditions). Moreover, these distinct regional isotopic signatures in lake water deviate markedly from the δ13C compositions of potential carbon sources (e.g., atmospheric CO2, soil CO2, methane oxidation CO2, and carbonates), indicating that δ13C of inorganic carbon can be applied for quantifying carbon source contributions to lakes in global carbon cycle studies.
Wildfire is a dominant disturbance in terrestrial ecosystems and a major driver of interannual variability in the global carbon cycle. However, the global wildfire carbon footprint remains unquantified due to uncertainties arising from its vast scale and variable fire behavior and fuels, thus complicating our understanding of the global carbon cycle. Using three satellite-derived wildfire carbon-emission (WCE) datasets, we quantify the global and regional trends in WCE from 2003 to 2024 and identify emerging hotspots of fire-carbon feedback risk. Global WCE averaged 8.56±0.97 Pg CO2 yr−1 and exhibited a significant long-term decline (−125±20 Tg CO2 yr−2, p<0.01), though trends varied sharply across regions and ecosystems. Africa had the steepest decline (−77.2±9.5 Tg CO2 yr−2, p<0.01), while North America showed an increasing trend (12.1±5.9 Tg CO2 yr−2). China also experienced a significant decline (−3.0±0.5 Tg CO2 yr−2, p<0.01) driven primarily by reduced forest-fire emissions. Across biomes, grasslands exhibited the strongest decrease (−112.1±9.5 Tg CO2 yr−2, p<0.01), whereas boreal forests showed a modest but concerning rise (6.5±5.9 Tg CO2 yr−2). Approximately 2.33×107 km2 (22.4
China has experienced a significant increase in dust storm frequency in recent years, yet achieving accurate forecasts for these events remains challenging. This study pioneers a targeted observation approach designed to elevate meteorological forecasting capabilities for dust storms, with the ultimate aim of achieving precise dust storm prediction. By analyzing forecast uncertainties in meteorological conditions for 2021–2024 Mongolian Cyclone-type dust storm events in the Beijing-Tianjin-Hebei (BTH) region using the WRF model, we propose a novel framework that holds potential for improving dust storm forecasting accuracy through an optimized observation strategy. The breakthrough lies in establishing a Meteorological Phase-Dust Dynamics (MP-DD) linkage via the Conditional Nonlinear Optimal Perturbation method, which redefines meteorological observation priority areas strategically located in the upwind area north of high dust concentration regions. This finding challenges the conventional singular vector method, which prioritizes meteorological field accuracy within high dust concentration regions as the key forecasting factor. The MP-DD framework not only provides a novel physics-based optimization strategy for targeted meteorological observations associated with dust storm forecasts but also underscores the critical need for China-Mongolia collaboration to combat cross-border environmental risks.
