Qingsong Zhang

Determining the distribution of soil organic carbon (SOC) in subsoil (20–100 cm depth) is important with respect to the global C cycle and warming mitigation. However, significant knowledge gaps remain regarding the spatiotemporal dynamics of SOC within this layer. By integrating traditional depth functions with machine learning approaches, we quantified soil β values, which represent the relative rate of decline in SOC density with depth, and provided high-resolution assessments of SOC dynamics across global ecosystems, including cropland, grassland, and forestland. The estimated subsoil SOC densities were 62 Mg ha⁻¹ (95 % CI: 52–73) for cropland, 70 Mg ha⁻¹ (95 % CI: 57–83) for grassland, and 97 Mg ha⁻¹ (95 % CI: 80–117) for forestland. SOC density exhibited a consistent decline with depth, ranging from 30 to 5 Mg ha⁻¹ in cropland, 32 to 7 Mg ha⁻¹ in grassland, and 40 to 13 Mg ha⁻¹ in forestland, across 20 cm depth increments from 20 to 100 cm. The estimated global subsoil SOC stock was 803 Pg C, with cropland, grassland, and forestland contributing 74, 181, and 547 Pg C, respectively. On average, 57 % of this carbon was stored within the top 0–100 cm of the soil profile. This study provides information on the vertical distribution and spatial patterns of SOC density at a 10 km resolution across global ecosystems, providing a scientific basis for future studies pertaining to Earth system models. The dataset is open-access and is available at https://doi.org/10.5281/zenodo.15019078 (Wang et al., 2025).

Determining the dynamics of organic carbon in subsoil (SOC, depth of 20–100 cm) is important with respect to the global C cycle and warming mitigation. However, there is still a huge knowledge gap in the dynamics of spatiotemporal changes in SOC in this layer. Combining traditional depth functions and machine-learning methods, we achieved soil β values and SOC dynamics at high resolution for global ecosystems (cropland, grassland, and forestland). First, quantified the spatial variability characteristics of soil β values and driving factors by analyzing 1221 soil profiles (0–100 cm) of globally distributed field observations. Then, based on multiple environmental variables and soil profile data, we mapped the grid-level soil β values with machine-learning approaches. Lastly, we evaluated the SOC density spatial distribution in different soil layers to determine the subsoil SOC stocks of various ecosystems. The subsoil SOC density values of cropland, grassland, and forestland were 63.8, 83.3, and 100.4 Mg ha^–1, respectively. SOC density decreased with increasing depth, ranging from 5.6 to 30.8 Mg ha^–1 for cropland, 7.5 to 40.0 Mg ha^–1 for grassland, and 9.6 to 47.0 Mg ha^–1 for forestland. The global subsoil SOC stock was 912 Pg C (cropland, grassland, and forestland were 67, 200, and 644 Pg C), in which an average of 54 % resided in the top 0–100 cm of the soil profile. Our results provide information on the vertical distribution and spatial patterns of SOC density at a 10 km resolution for areas of Global ecosystems, which providing a scientific basis for future studies pertaining to Earth system models. The dataset is open-access and available at https://doi.org/10.5281/zenodo.10846543 (Wang et al., 2024).

Researchers have previously described the response of crop productivity and greenhouse gas (GHG) emissions to fertilizer nitrogen (N) additions, but they have not determined how to maximize yields while minimizing GHG emissions. We conducted an experiment at 2293 sites with four N levels to simulate both grain yield and yield-scaled GHG emissions in response to the N addition. The yield-scaled GHG emissions decreased by 16% as the N rate increased from treatments without the N addition to the minimum yield-scaled GHG emissions, which was comparable to the values associated with the maximum grain yields. The sites with both high soil productivity and high crop productivity had the highest yield and lowest yield-scaled GHG emissions, with 43% higher yield and 38% lower yield-scaled GHG emissions than sites with low soil and low crop productivity. These findings are expected to enhance evaluations of wheat production and GHG emissions in China, and thereby contribute to addressing disparities in the global food and GHG budget.

China’s rapid increase in mass excreta and its environmental discharge have attracted substantial attention. However, cropland as a main destination of excreta utilization has not been extensively evaluated. Here, a national survey was used to assess the utilization of manure in croplands across China. The data included the inputs of manure nitrogen (N), phosphorus (P), and potassium (K) for cereals, fruits, vegetables, and other crops, along with the manure proportion of total N, P, and K inputs at the county level. The results showed that the manure N, P, and K inputs were 6.85, 2.14, and 4.65 million tons (Mt), respectively, constituting 19.0%, 25.5%, and 31.1% of the total N, P, and K, respectively. The spatial distribution of the manure proportion of total inputs was lower in Eastern China and higher in Western China. The results provide a detailed description of the utilization of manure nutrients in agricultural areas throughout China, which will serve as basic support for policymakers and researchers involved in future agricultural nutrient management in China.

China feeds 19.1% of the world’s population with 8.6% of the arable land. Here we propose an integrated approach combining crop redistribution and improved management to meet China’s food demand in 2030. We simulated the food demand, estimated the national crop production through the productivity of the top 10% of producers in each county, and optimized the spatial distribution of 11 groups of crop types among counties using the data of the top producers. Integrating crop redistribution and improved management increased crop production and can meet the food demand in 2030, while the agricultural inputs (N and P fertilizers and irrigation water) and environmental impacts (reactive N loss and greenhouse gas emissions) were reduced. Although there are significant socio-economic and cultural barriers to implementing such redistribution, these results suggest that integrated measures can achieve food security and decrease negative environmental impacts. County-specific policies and advisory support will be needed to achieve the promises of combining optimization strategies.

Excessive use of synthetic nitrogen (N) for Chinese wheat production results in high loss of reactive N loss (Nr; all forms of N except N₂) into the environment, causing serious environmental issues. Quantifying Nr loss and its spatial variations therein is vital to optimize N management and mitigate loss. However, accurate, high spatial resolution estimations of Nr from wheat production are lacking due to limitations of data generation and estimation methods. Here, we applied the random forest (RF) algorithm to bottom-up N application rate data, obtained through a survey of millions of farmers, to estimate the Nr loss from wheat production in 2014. The results showed that the average total Nr loss was 52.5 kg N ha⁻¹ (range: 4.6-157.8 kg N ha⁻¹), which accounts for 26.1% of the total N applied. The hotspots for high Nr loss are the same as those high applied N, including northwestern Xinjiang, central-southern Hebei, Shandong, central-northern Jiangsu, and Hubei. Our database could guide regional N management and be used in conjunction with biogeochemical models.

Sustainably feeding the growing population in China attracts attention globally. Despite practices success, producing enough food to simultaneously address resource and pollution problems has been infeasible. To assess how to achieve this goal in 2035, we created a pathway that synergistic combining improved managements and cropland redistribution based on 11.1 million farmer surveys and 4,272 georeferenced field observations. Here, we firstly selected the practices of top 10% performers in crop yield and nitrogen (N) efficiency as crop-specific attainable improved managements at the county level. The optimized crop distribution within improved managements was then performed to minimize inputs (N and phosphorus fertilizer, irrigation water) or environmental impacts (reactive N [Nr] loss and greenhouse gas [GHG] emissions). We identified that combing improved managements and cropland redistribution could produce enough food demands in 2035, with 24% more production compared to 2012. It also reduced the inputs and environmental impacts in a range of 19%-35%, mainly sourced from the central and eastern coastal areas by improved productivity and diminished cropland of fruit and vegetables. These findings highlight the necessity for a synergistic combination of measures to sustainably feed the growing population and establish a more realistic and effective policy.

Growing evidence indicates that climate change has significant impacts on crop production. However, the spatial variability and extent of the impacts on N concentrations in crops have yet to be explicitly described, although N concentration is a critical element of both the nutritional value of food and the planning of appropriate field management procedures. We present multi‐tiered evidence showing that warming increases N concentration in rice and reduces internal N use efficiency (iNUE), thereby increasing N uptake in crops and encouraging greater N fertilizer use. First, we mapped current N concentrations and iNUE for single‐season irrigation paddy rice production in 975 counties across China using data from 1,637 on‐farm experiments and Random Forest regression modeling. Current N concentrations in rice grain and straw have increased with increasing temperatures from northern to southern China. The global meta‐analysis provided further direct evidence that warming increases N concentration in rice. Future warming projections indicate that N concentrations in grain will increase by 5%–9.6% and that iNUE will decrease by 9%–14% under scenarios RCP4.5 and RCP 8.5 although higher CO₂ and other factors may also affect N concentrations. Overall, these data demonstrate that N uptake by rice will need to increase by 7.8%–13.6% to maintain current grain yields across China based on warming alone. Our findings indicate that warming associated with increase in N concentration and reduction in iNUE would be expected to encourage N application, raise the urgency of measures to prevent excessive N application, and stronger counteracting measures need to be enacted.

China purchases around 66% of the soy that is traded internationally. This strains the global food supply and contributes to greenhouse gas emissions. Here we show that optimizing the maize and soy production of China can improve its self-sufficiency and also alleviate adverse environmental effects. Using data from more than 1,800 counties in China, we estimate the area-weighted yield potential (Y_pot) and yield gaps, setting the attainable yield (Y_att) as the yield achieved by the top 10% of producers per county. We also map out county-by-county acreage allocation and calculate the attainable production capacity according to a set of sustainability criteria. Under optimized conditions, China would be able to produce all the maize and 45% of the soy needed by 2035—while reducing nitrogen fertilizer use by 26%, reactive nitrogen loss by 28% and greenhouse gas emissions by 19%—with the same acreage as 2017, our reference year.