Study on the multifunctional spatial-temporal evolution and coupling coordination of cultivated land: A case study of Hebei Province, China.

The rational use of cultivated land can guarantee food security and thus is highly important for ensuring social stability, economic development and national security. The current study investigated the multifunctional temporal and spatial variation characteristics of cultivated land and explored the spatial and temporal characteristics of the multifunction and coupling coordination degrees of cultivated land throughout Hebei Province. Based on the administrative division data, statistical yearbook data and land use status data of the impacted areas, a multifunctional evaluation index system of cultivated land was established. The CRITIC weight method and entropy weight method were used to determine the weight of the index, the comprehensive index model was used to determine the production, social security, ecology and landscape functions of cultivated land of Hebei Province in different periods, the coupling coordination model was used to explore the multifunctional coupling coordination degree of cultivated land in each county, and spatial autocorrelation analysis was performed to determine the correlation of the multifunctional coupling coordination degrees. From 2000 to 2020, the production, social security and landscape function of cultivated land in Hebei Province trended upward; the ecological function trended slightly downward. The multifunctional coupling coordination degree of cultivated land in Hebei Province trended significantly upward and changed from limited coordination to intermediate coordination. Furthermore, it exhibited strong agglomeration and a significant positive spatial correlation, forming a 'V'-type change rule of first decreasing and then increasing. Hebei Province exhibited remarkable spatial and temporal characteristics of the multifunction and coupling coordination degrees of cultivated land. Regions could thus customize different cultivated land functions to maximize the benefits of cultivated land use. The findings of this study may provide a scientific basis and theoretical support for sustainably using and managing cultivated land resources in areas with similar human geographical environments.

Multi-criteria GIS-based land suitability analysis for rice cultivation: a case study in Guilan Province, Iran.

Ensuring food security and sustainable resource management has become a paramount global concern, prompting significant attention to land suitability analysis for enhancing agricultural production. In this study, an AHP-weighted overlay method was employed to delineate rice cultivation suitability in Guilan province, Iran, a central hub for rice production. Sixteen climatic, topographic, and soil variables were integrated, and individual maps were reclassified to align with the specific requisites for rice production. The results revealed three suitability classes: including 'very suitable,' 'suitable,' and 'moderately suitable', covering 91%, 6%, and 3% of the land, respectively. Soil attributes, particularly organic matter, significantly influenced suitability (weight value of 0.745), with topographic and soil factors outweighing climate in assessment. While salinity is generally absent, organic matter deficiency affects 44% of the land. Phosphorus imbalances are prevalent, with potassium toxicity observed in 10%. Microelement deficiencies, especially in iron and zinc, are noted. Additionally, the results indicated that topographic and soil attributes played a more significant role than climate-related factors in assessing land suitability for rice cultivation within the study area. This research provides a comprehensive spatial analysis of all variables in the study region, shedding light on the complexities of land suitability for rice cultivation. These findings contribute to the understanding of agricultural sustainability and resource management strategies in the context of food security.

The potential climate benefits of seaweed farming in temperate waters.

Seaweed farming is widely promoted as an approach to mitigating climate change despite limited data on carbon removal pathways and uncertainty around benefits and risks at operational scales. We explored the feasibility of climate change mitigation from seaweed farming by constructing five scenarios spanning a range of industry development in coastal British Columbia, Canada, a temperate region identified as highly suitable for seaweed farming. Depending on growth rates and the fate of farmed seaweed, our scenarios sequestered or avoided between 0.20 and 8.2 Tg CO2e year-1, equivalent to 0.3% and 13% of annual greenhouse gas emissions in BC, respectively. Realisation of climate benefits required seaweed-based products to replace existing, more emissions-intensive products, as marine sequestration was relatively inefficient. Such products were also key to reducing the monetary cost of climate benefits, with product values exceeding production costs in only one of the scenarios we examined. However, model estimates have large uncertainties dominated by seaweed production and emissions avoided, making these key priorities for future research. Our results show that seaweed farming could make an economically feasible contribute to Canada's climate goals if markets for value-added seaweed based products are developed. Moreover, our model demonstrates the possibility for farmers, regulators, and researchers to accurately quantify the climate benefits of seaweed farming in their regional contexts.

A critical systematic review on spectral-based soil nutrient prediction using machine learning.

The United Nations (UN) emphasizes the pivotal role of sustainable agriculture in addressing persistent starvation and working towards zero hunger by 2030 through global development. Intensive agricultural practices have adversely impacted soil quality, necessitating soil nutrient analysis for enhancing farm productivity and environmental sustainability. Researchers increasingly turn to Artificial Intelligence (AI) techniques to improve crop yield estimation and optimize soil nutrition management. This study reviews 155 papers published from 2014 to 2024, assessing the use of machine learning (ML) and deep learning (DL) in predicting soil nutrients. It highlights the potential of hyperspectral and multispectral sensors, which enable precise nutrient identification through spectral analysis across multiple bands. The study underscores the importance of feature selection techniques to improve model performance by eliminating redundant spectral bands with weak correlations to targeted nutrients. Additionally, the use of spectral indices, derived from mathematical ratios of spectral bands based on absorption spectra, is examined for its effectiveness in accurately predicting soil nutrient levels. By evaluating various performance measures and datasets related to soil nutrient prediction, this paper offers comprehensive insights into the applicability of AI techniques in optimizing soil nutrition management. The insights gained from this review can inform future research and policy decisions to achieve global development goals and promote environmental sustainability.

Redesigning and validation of fertilizer use in maize for variable plant densities in central rift valley and Jimma in Ethiopia.

Due to low adoption and sub-optimal fertilizer use and planting density recommendation in maize, redesigning and testing these technologies are required. The study was conducted to evaluate redesigned fertilizer use of maize in two pant densities (32,443 and 53,333 plants ha-1 in Central Rift Valley (CRV); 27724 and 62,000 plants ha-1 in Jimma) on farmers' fields in contrasting agro-ecologies of Ethiopia. The on-farm study was conducted in the 2017 and 2018 cropping seasons with 3 × 2 fertilizer and plant density, factors in both regions of Ethiopia. In redesigned fertilizer use, nutrients were estimated based on the target yield. In this study, 40.8, 0.0, and 12.2 kg ha-1 N, P, and K were estimated for the redesigned fertilizer use in CRV (50% of water-limited potential yield (Yw) = 3.1 t ha-1) whereas in Jimma (50% of Yw = 7.5 t ha-1) 149.8, 9, 130.6 kg ha-1 N, P and K were estimated to produce the 50% of Yw. Linear mixed modeling was used to assess the effect of fertilizer-plant density treatments on maize yield and nutrient use efficiency. The result revealed that the average estimated maize yield for WOF, FFU, and RDFU fertilizer treatments were 2.6, 3.6, and 4.5 t ha-1 under current plant density (32,443 plants ha-1) in CRV whereas the average yields of these treatments were 3.2, 4.5 and 4.5 t ha-1 respectively when maize was grown with redesigned plant density (53,333 plants ha-1) in the same location. The average maize yield with WOF, FFU, and RDFU were 3.0, 4.6, and 4.6 t ha-1 with 27,774 plants ha-1 plant density in Jimma whereas the average maize yields over the two seasons with the same treatments were 4.3, 6.0 and 8.0 t ha-1 respectively when the crop is planted with 62,000 plants ha-1 plant density. The RDFU and redesigned plant density resulted in significantly higher yield compared to their respective control CRV but RDFU significantly increased maize yield when it was planted at redesigned (62,000 plant ha-1) in Jimma. FFU and RDFU were economically viable and redesigned plant density was also a cheaper means of improving maize productivity, especially in the Jimma region. Soil organic carbon and N were closely related to the grain yield response of maize compared to other soil factors. In conclusion, this investigation gives an insight into the importance of redesigned fertilizer use and redesigned plant density for improving maize productivity and thereby narrowing the yield gaps of the crop in high maize potential regions in Ethiopia like Jimma.

Enhancing aromatic rice production through agronomic and nutritional management for improved yield and quality.

To meet the growing international demand for aromatic rice, this study, conducted at Uttar Banga Krishi Viswavidyalaya in Cooch Behar, West Bengal, aimed to enhance the yield and quality of the 'Tulaipanji' rice cultivar through advanced establishment methods and the use of organic nutrients over two years. The research tested three planting techniques: mechanical transplanting, wet direct seeding (using a drum seeder), and traditional methods, alongside four nutrient management strategies: vermicompost, farmyard manure, a mix of both, and conventional fertilizers. Findings revealed that mechanical transplanting significantly increased yield by over 31.98% and 71.05% compared to traditional methods and wet direct seeding, respectively. Using vermicompost alone as a nutrient source not only boosted yields by 21.31% over conventional fertilizers but also enhanced the rice's nutritional value and cooking quality. Moreover, soils treated with vermicompost showed higher dehydrogenase activity, indicating better soil health. Economically, mechanical transplanting with vermicompost was the most beneficial, yielding the highest net returns and benefit-cost ratios in both years studied. This approach presents a viable model for improving the sustainability of aromatic rice production globally, emphasizing the economic and environmental advantages of adopting mechanical planting techniques and organic fertilization methods.

Agronomic performance and technological quality of sugarcane submitted to different poultry litter dosages.

Sugarcane is a central crop for sugar and ethanol production. Investing in sustainable practices can enhance productivity, technological quality, mitigate impacts, and contribute to a cleaner energy future. Among the factors that help increase the productivity of sugarcane, the physical, chemical and biological parameters of the soil are amongst the most important. The use of poultry litter has been an important alternative for soil improvement, as it acts as a soil conditioner. Therefore, this work aimed to verify the best doses of poultry litter for the vegetative, reproductive and technological components of sugarcane. The experiment was carried out at Usina Denusa Destilaria Nova União S/A in the municipality of Jandaia, GO. The experimental design used was a complete randomized block design with four replications: 5 × 4, totaling 20 experimental units. The evaluated factor consisted of four doses of poultry litter plus the control (0 (control), 2, 4, 6 and 8 t ha-1). In this study, were evaluated the number of tillers, lower stem diameter, average stem diameter, upper stem diameter, plant height, stem weight and productivity. The technological variables of total recoverable sugar, recoverable sugar, Brix, fiber, purity and percentage of oligosaccharides were also evaluated. It was observed, within the conditions of this experiment, that the insertion of poultry litter did not interfere significantly in most biometric, productive and technological variables of the sugarcane. But it can also be inferred that there was a statistical trend toward better results when the sugarcane was cultivated with 4 t ha-1 of poultry litter.

Vulnerability assessment of aerosol and climate variability for rice and maize yield using EO datasets for sustainable agriculture over India.

Human-generated aerosol pollution gradually modifies the atmospheric chemical and physical attributes, resulting in significant changes in weather patterns and detrimental effects on agricultural yields. The current study assesses the loss in agricultural productivity due to weather and anthropogenic aerosol variations for rice and maize crops through the analysis of time series data of India spanning from 1998 to 2019. The average values of meteorological variables like maximum temperature (TMAX), minimum temperature (TMIN), rainfall, and relative humidity, as well as aerosol optical depth (AOD), have also shown an increasing tendency, while the average values of soil moisture and fraction of absorbed photosynthetically active radiation (FAPAR) have followed a decreasing trend over that period. This study's primary finding is that unusual variations in weather variables like maximum and minimum temperature, rainfall, relative humidity, soil moisture, and FAPAR resulted in a reduction in rice and maize yield of approximately (2.55%, 2.92%, 2.778%, 4.84%, 2.90%, and 2.82%) and (5.12%, 6.57%, 6.93%, 6.54%, 4.97%, and 5.84%), respectively. However, the increase in aerosol pollution is also responsible for the reduction of rice and maize yield by 7.9% and 8.8%, respectively. In summary, the study presents definitive proof of the detrimental effect of weather, FAPAR, and AOD variability on the yield of rice and maize in India during the study period. Meanwhile, a time series analysis of rice and maize yields revealed an increasing trend, with rates of 0.888 million tons/year and 0.561 million tons/year, respectively, due to the adoption of increasingly advanced agricultural techniques, the best fertilizer and irrigation, climate-resilient varieties, and other factors. Looking ahead, the ongoing challenge is to devise effective long-term strategies to combat air pollution caused by aerosols and to address its adverse effects on agricultural production and food security.

Advancing common bean (Phaseolus vulgaris L.) disease detection with YOLO driven deep learning to enhance agricultural AI.

Common beans (CB), a vital source for high protein content, plays a crucial role in ensuring both nutrition and economic stability in diverse communities, particularly in Africa and Latin America. However, CB cultivation poses a significant threat to diseases that can drastically reduce yield and quality. Detecting these diseases solely based on visual symptoms is challenging, due to the variability across different pathogens and similar symptoms caused by distinct pathogens, further complicating the detection process. Traditional methods relying solely on farmers' ability to detect diseases is inadequate, and while engaging expert pathologists and advanced laboratories is necessary, it can also be resource intensive. To address this challenge, we present a AI-driven system for rapid and cost-effective CB disease detection, leveraging state-of-the-art deep learning and object detection technologies. We utilized an extensive image dataset collected from disease hotspots in Africa and Colombia, focusing on five major diseases: Angular Leaf Spot (ALS), Common Bacterial Blight (CBB), Common Bean Mosaic Virus (CBMV), Bean Rust, and Anthracnose, covering both leaf and pod samples in real-field settings. However, pod images are only available for Angular Leaf Spot disease. The study employed data augmentation techniques and annotation at both whole and micro levels for comprehensive analysis. To train the model, we utilized three advanced YOLO architectures: YOLOv7, YOLOv8, and YOLO-NAS. Particularly for whole leaf annotations, the YOLO-NAS model achieves the highest mAP value of up to 97.9% and a recall of 98.8%, indicating superior detection accuracy. In contrast, for whole pod disease detection, YOLOv7 and YOLOv8 outperformed YOLO-NAS, with mAP values exceeding 95% and 93% recall. However, micro annotation consistently yields lower performance than whole annotation across all disease classes and plant parts, as examined by all YOLO models, highlighting an unexpected discrepancy in detection accuracy. Furthermore, we successfully deployed YOLO-NAS annotation models into an Android app, validating their effectiveness on unseen data from disease hotspots with high classification accuracy (90%). This accomplishment showcases the integration of deep learning into our production pipeline, a process known as DLOps. This innovative approach significantly reduces diagnosis time, enabling farmers to take prompt management interventions. The potential benefits extend beyond rapid diagnosis serving as an early warning system to enhance common bean productivity and quality.

Tripartite evolutionary game and simulation analysis of agricultural non-point source pollution control.

Agricultural non-point source pollution control (ANSPC) is a complex, long-term and dynamic environmental protection process. In order to motivate multiple subjects to participate in ANSPC, this paper constructs a tripartite evolutionary game model of local government, village collectives and farmers, which explores the strategic choices and influencing factors of different subjects through simulation analysis. The results indicate that: There are five stable strategy points in the ANSPC game system, which can be divided into four stages based on subject interactions. Village collectives should play an intermediary role in ANSPC and try to coordinate the behaviour of different subjects. The ideal and stable evolution state is "weak supervise, positive response, and active participate", but it cannot be realized at present. The strategy selection of subjects is determined by relative net income. Providing penalties requires considering the heterogeneity of subjects, but incentives are beneficial for achieving tripartite governance. This study provides new evidence for understanding the role of multi-agency participation in agricultural non-point source pollution control, and provides theoretical guidance for the government to formulate differentiated intervention mechanisms, which is an important reference for achieving sustainable development goals.

Deep learning-based prediction of plant height and crown area of vegetable crops using LiDAR point cloud.

Remote sensing has been increasingly used in precision agriculture. Buoyed by the developments in the miniaturization of sensors and platforms, contemporary remote sensing offers data at resolutions finer enough to respond to within-farm variations. LiDAR point cloud, offers features amenable to modelling structural parameters of crops. Early prediction of crop growth parameters helps farmers and other stakeholders dynamically manage farming activities. The objective of this work is the development and application of a deep learning framework to predict plant-level crop height and crown area at different growth stages for vegetable crops. LiDAR point clouds were acquired using a terrestrial laser scanner on five dates during the growth cycles of tomato, eggplant and cabbage on the experimental research farms of the University of Agricultural Sciences, Bengaluru, India. We implemented a hybrid deep learning framework combining distinct features of long-term short memory (LSTM) and Gated Recurrent Unit (GRU) for the predictions of plant height and crown area. The predictions are validated with reference ground truth measurements. These predictions were validated against ground truth measurements. The findings demonstrate that plant-level structural parameters can be predicted well ahead of crop growth stages with around 80% accuracy. Notably, the LSTM and the GRU models exhibited limitations in capturing variations in structural parameters. Conversely, the hybrid model offered significantly improved predictions, particularly for crown area, with error rates for height prediction ranging from 5 to 12%, with deviations exhibiting a more balanced distribution between overestimation and underestimation This approach effectively captured the inherent temporal growth pattern of the crops, highlighting the potential of deep learning for precision agriculture applications. However, the prediction quality is relatively low at the advanced growth stage, closer to the harvest. In contrast, the prediction quality is stable across the three different crops. The results indicate the presence of a robust relationship between the features of the LiDAR point cloud and the auto-feature map of the deep learning methods adapted for plant-level crop structural characterization. This approach effectively captured the inherent temporal growth pattern of the crops, highlighting the potential of deep learning for precision agriculture applications.

Impact of varied tillage practices and phosphorus fertilization regimes on wheat yield and grain quality parameters in a five-year corn-wheat rotation system.

Choosing appropriate tillage methods and applying the right amount of chemical fertilizers are pivotal for optimizing wheat management and enhancing wheat quality. This study investigated the influence of conservation agriculture and phosphorus levels on nutrient content, yield components, and quality traits of wheat in a corn-wheat rotation. Conducted over five years in field conditions, the study employed a randomized complete block design with tillage treatments (conventional tillage, CT; minimum tillage, MT; and no tillage, NT) and phosphorus levels (no fertilizer use, P0; and 100% fertilizer recommendation, PR) as factors. Soil samples were collected during the fourth year (2021-2022). Results revealed significant impacts of tillage methods and phosphorus levels on wheat straw and grain nutrient composition, yield components, and quality traits. Conventional tillage yielded the highest values for protein content (12%), Zeleny sedimentation volume (20.33 mL), hardness index (45), water absorption (64.12%), and wet gluten content (25.83%). Additionally, phosphorus fertilizer application positively influenced protein percentage, gluten weight, and gluten index. The study highlights the potential of strategic soil management, particularly conventional tillage combined with phosphorus fertilization, to enhance wheat quality and yield. By elucidating these relationships, the findings contribute to optimizing wheat cultivation practices and advancing the development of superior wheat cultivars for baking applications.

Biocalorimetry-aided monitoring of fungal pretreatment of lignocellulosic agricultural residues.

The present study aimed to investigate whether and how non-invasive biocalorimetric measurements could serve for process monitoring of fungal pretreatment during solid-state fermentation (SSF) of lignocellulosic agricultural residues such as wheat straw. Seven filamentous fungi representing different lignocellulose decay types were employed. Water-soluble sugars being immediately available after fungal pretreatment and those becoming water-extractable after enzymatic digestion of pretreated wheat straw with hydrolysing (hemi)cellulases were considered to constitute the total bioaccessible sugar fraction. The latter was used to indicate the success of pretreatments and linked to corresponding species-specific metabolic heat yield coefficients (YQ/X) derived from metabolic heat flux measurements during fungal wheat straw colonisation. An YQ/X range of about 120 to 140 kJ/g was seemingly optimal for pretreatment upon consideration of all investigated fungi and application of a non-linear Gaussian fitting model. Upon exclusion from analysis of the brown-rot basidiomycete Gloeophyllum trabeum, which differs from all other here investigated fungi in employing extracellular Fenton chemistry for lignocellulose decomposition, a linear relationship where amounts of total bioaccessible sugars were suggested to increase with increasing YQ/X values was obtained. It remains to be elucidated whether an YQ/X range being optimal for fungal pretreatment could firmly be established, or if the sugar accessibility for post-treatment generally increases with increasing YQ/X values as long as "conventional" enzymatic, i.e. (hemi)cellulase-based, lignocellulose decomposition mechanisms are operative. In any case, metabolic heat measurement-derived parameters such as YQ/X values may become very valuable tools supporting the assessment of the suitability of different fungal species for pretreatment of lignocellulosic substrates. KEY POINTS: • Biocalorimetry was used to monitor wheat straw pretreatment with seven filamentous fungi. • Metabolic heat yield coefficients (YQ/X) seem to indicate pretreatment success. • YQ/X values may support the selection of suitable fungal strains for pretreatment.

Landscape features support natural pest control and farm income when pesticide application is reduced.

Future trajectories of agricultural productivity need to incorporate environmental targets, including the reduction of pesticides use. Landscape features supporting natural pest control (LF-NPC) offer a nature-based solution that can serve as a partial substitute for synthetic pesticides, thereby supporting future productivity levels. Here, we introduce a novel approach to quantify the contribution of LF-NPC to agricultural yields and its associated economic value to crop production in a broad-scale context. Using the European Union as case study, we combine granular farm-level data, a spatially explicit map of LF-NPC potential, and a regional agro-economic supply and market model. The results reveal that farms located in areas characterized by higher LF-NPC potential experience lower productivity losses in a context of reduced synthetic pesticides use. Our analysis suggests that LF-NPC reduces yield gaps on average by four percentage points, and increases income by a similar magnitude. These results highlight the significance of LF-NPC for agricultural production and income, and provide a valuable reference point for farmers and policymakers aiming to successfully invest in landscape features to achieve pesticides reduction targets.

Belowground plant allocation regulates rice methane emissions from degraded peat soils.

Carbon-rich peat soils have been drained and used extensively for agriculture throughout human history, leading to significant losses of their soil carbon. One solution for rewetting degraded peat is wet crop cultivation. Crops such as rice, which can grow in water-saturated conditions, could enable agricultural production to be maintained whilst reducing CO2 and N2O emissions from peat. However, wet rice cultivation can release considerable methane (CH4). Water table and soil management strategies may enhance rice yield and minimize CH4 emissions, but they also influence plant biomass allocation strategies. It remains unclear how water and soil management influences rice allocation strategies and how changing plant allocation and associated traits, particularly belowground, influence CH4-related processes. We examined belowground biomass (BGB), aboveground biomass (AGB), belowground:aboveground ratio (BGB:ABG), and a range of root traits (root length, root diameter, root volume, root area, and specific root length) under different soil and water treatments; and evaluated plant trait linkages to CH4. Rice (Oryza sativa L.) was grown for six months in field mesocosms under high (saturated) or low water table treatments, and in either degraded peat soil or degraded peat covered with mineral soil. We found that BGB and BGB:AGB were lowest in water saturated conditions where mineral soil had been added to the peat, and highest in low-water table peat soils. Furthermore, CH4 and BGB were positively related, with BGB explaining 60% of the variation in CH4 but only under low water table conditions. Our results suggest that a mix of low water table and mineral soil addition could minimize belowground plant allocation in rice, which could further lower CH4 likely because root-derived carbon is a key substrate for methanogenesis. Minimizing root allocation, in conjunction with water and soil management, could be explored as a strategy for lowering CH4 emissions from wet rice cultivation in degraded peatlands.

Opportunities to produce food from substantially less land.

The vast majority of the food we eat comes from land-based agriculture, but recent technological advances in agriculture and food technology offer the prospect of producing food using substantially less or even virtually no land. For example, indoor vertical farming can achieve very high yields of certain crops with a very small area footprint, and some foods can be synthesized from inorganic precursors in industrial facilities. Animal-based foods require substantial land per unit of protein or per calorie and switching to alternatives could reduce demand for some types of agricultural land. Plant-based meat substitutes and those produced through fermentation are widely available and becoming more sophisticated while in the future cellular agricultural may become technically and economical viable at scale. We review the state of play of these potentially disruptive technologies and explore how they may interact with other factors, both endogenous and exogenous to the food system, to affect future demand for land.

How can the digital economy reduce carbon emissions? Empirical evidence from China.

China is transitioning into the digital economy era. The advancement of the digital economy could offer a fresh mechanism to attain carbon peak and carbon neutrality objectives. Applications of the digital economy, such as smart energy management, intelligent transport systems, and digital agricultural technologies, have significantly reduced carbon emissions by optimizing resource use, reducing energy waste, and improving production efficiency. This research does so by devising a theoretical model that looks into the multi-faceted power of the digital economy under a two-sector paradigm. Utilising a panel model, a mediation effect model and a spatial Durbin model to assess the digital economy's power on carbon emissions. This research has determined that the digital economy can significantly diminish carbon emissions, with green tech innovations and industrial transformation being key contributors. The spatial spillover effect was used for the digital economy to aid in lowering carbon emissions in adjacent districts and upgrading better environmental stewardship. The influence of the digital economy has better performance in lowering carbon emissions in mid-western China than in the eastern area. This paper deepens understanding of the drivers of low-carbon growth and the significance, mechanism and regional disparities of the digital economy's effect on reducing carbon emissions. It offers valuable policy insights and guidance for globally achieving digital economy growth, reducing carbon emissions and reaching carbon peak and neutrality goals.

Assessing the impact of climate variability on maize yields in the different regions of Ghana-A machine learning perspective.

Climate variability has become one of the most pressing issues of our time, affecting various aspects of the environment, including the agriculture sector. This study examines the impact of climate variability on Ghana's maize yield for all agro-ecological zones and administrative regions in Ghana using annual data from 1992 to 2019. The study also employs the stacking ensemble learning model (SELM) in predicting the maize yield in the different regions taking random forest (RF), support vector machine (SVM), gradient boosting (GB), decision tree (DT), and linear regression (LR) as base models. The findings of the study reveal that maize production in the regions of Ghana is inconsistent, with some regions having high variability. All the climate variables considered have positive impact on maize yield, with a lesser variability of temperature in the Guinea savanna zones and a higher temperature variability in the Volta Region. Carbon dioxide (CO2) also plays a significant role in predicting maize yield across all regions of Ghana. Among the machine learning models utilized, the stacking ensemble model consistently performed better in many regions such as in the Western, Upper East, Upper West, and Greater Accra regions. These findings are important in understanding the impact of climate variability on the yield of maize in Ghana, highlighting regional disparities in maize yield in the country, and highlighting the need for advanced techniques for forecasting, which are important for further investigation and interventions for agricultural planning and decision-making on food security in Ghana.

Measurement and spatio-temporal heterogeneity analysis of coupling coordination between development of digital economy and agricultural carbon emission performance.

The new development pattern has identified two key avenues for the sustained advancement of high-quality agricultural and rural development: digitalisation and low-carbon development. The measurement of the digital economy and the agricultural carbon emission performance, and their spatial and temporal heterogeneity, is a crucial step in promoting the spatial coordination and sustainable development of digitalisation and low-carbon agriculture. This paper employs the entropy value method, SBM model, and coupling coordination degree model to investigate the coupling coordination measurement and spatial-temporal heterogeneity of the performance of the digital economy and agricultural carbon emissions. The data used are provincial panel data from 2013 to 2021. The simulation results demonstrate that, between 2013 and 2021, the digital economy of all provinces exhibited varying degrees of growth, yet the development of the digital economy between provinces exhibited a more pronounced tendency to diverge. Concurrently, the agricultural carbon emission efficiency in China exhibited a fluctuating upward trend. The development of the digital economy and the efficiency of agricultural carbon emission were found to be highly coupled. Their coupling and coordination relationship showed a downward trend followed by an upward trend. In general, it is suggested that we should increase investment in digital economy infrastructure and technology, promote digital agricultural applications, strengthen policy guidance and financial support, establish a coupling coordination mechanism and strengthen farmers' digital literacy and environmental awareness.

An optimized crop-livestock system can achieve a safe and just planetary boundary for phosphorus at the sub-basin level in China.

The contribution of crop and livestock production to the exceedance of the planetary boundary for phosphorus (P) in China is still unclear, despite the country's well-known issues with P fertilizer overuse and P-related water pollution. Using coupled models at sub-basin scales we estimate that livestock production increased the consumption of P fertilizer fivefold and exacerbated P losses twofold from 1980 to 2017. At present, China's crop-livestock system is responsible for exceeding what is considered a 'just' threshold for fertilizer P use by 30% (ranging from 17% to 68%) and a 'safe' water quality threshold by 45% (ranging from 31% to 74%) in 25 sub-basins in China. Improving the crop-livestock system will keep all sub-basins within safe water quality and just multigenerational limits for P in 2050.