Green high-yield and high-efficiency technology: a new path balancing yield and ecology
image:
Image
view moreCredit: Jian HUANG , Yixiao CHAI , Shichao YANG , Yiwen CAO , Lei YANG , Min WANG , Xusheng MENG , Shiwei GUO
As a staple food for more than half of the global population, the high and stable yield of rice is directly related to food security. As the world’s largest rice producer, China has increased rice yield per unit through intensive fertilization and flood irrigation, but this model has also brought problems such as soil degradation, water pollution, and greenhouse gas emissions. How to ensure food supply while breaking through resource and environmental constraints?
Xusheng Meng and colleagues from Nanjing Agricultural University proposed a green, high-yield, and high-efficiency rice technology system in a review study, providing a solution to this problem. The related paper has been published in Frontiers of Agricultural Science and Engineering (DOI: 10.15302/J-FASE-2025636).
Currently, China’s rice production is facing the dilemma of “high input and low efficiency”. Data shows that China accounts for 20% of the global rice planting area but consumes 37% of the nitrogen fertilizer, with nitrogen use efficiency lower than the world average. Excessive nitrogen fertilizer enters the environment through farmland runoff, leaching, and volatilization, leading to soil acidification and compaction, and exacerbating lake eutrophication and air pollution. At the same time, traditional flooded cultivation makes paddy fields an important source of greenhouse gas emissions. China’s paddy fields emit 712 million tons of carbon dioxide equivalents annually, higher than other major rice-producing countries.
In response to these challenges, researchers proposed three innovative technical paths based on experimental data from multiple regions across the country. The first is to optimize nutrient management strategies, balancing the rice population structure through the fertilization method of “reducing basal-tillering fertilizer and increasing panicle fertilizer”. In traditional cultivation, farmers tend to apply a large amount of fertilizer at the seedling stage to promote tillering, but excessive tillering easily leads to an increase in ineffective panicles and the risk of lodging. The new strategy precisely adjusts the proportion of nitrogen fertilizer allocated in different growth stages, which can not only promote effective tillering but also improve the development quality of panicles and grains in the later stage. Experiments show that this can increase nitrogen use efficiency by 8.1%–21.3%.
The second is the “carbon-nitrogen synergy” technology for improving soil fertility. The study found that combining crushed straw returning with organic fertilizer replacing part of chemical fertilizer can significantly increase soil organic carbon content and enhance the soil’s ability to retain water and nutrients. Long-term experiments show that this model can reduce ammonia volatilization loss by more than 17%, while activating the activity of functional microorganisms such as nitrogen-fixing bacteria and phosphate-solubilizing bacteria, and promoting nutrient conversion efficiency.
The third is the integrated water management technology of “water-saving and controlled drainage”. Different from the traditional full-period flooding, the “alternate wetting and drying” irrigation mode improves soil aeration, promotes root development, and reduces methane emissions by properly drying the fields in the late tillering stage. Demonstrations in the double-cropping rice areas of South China show that this technology can save 19% of water compared with conventional irrigation, reduce methane emissions by 16.2%, and keep the yield stable.
Researchers also proposed differentiated technical schemes according to the regional characteristics of China’s five major rice-growing regions. For example, in the Northeast region, nitrogen-zinc synergistic fertilization technology is used to solve the problem of seedling stunting caused by low temperature in early spring; in the mountainous areas of Southwest China, technologies such as sparse planting for strong plants and deep application of organic fertilizer are promoted to cope with topographical constraints; in the arid regions of Northwest China, film mulching hole sowing combined with controlled-release fertilizer is adopted to achieve water-saving and high yield. These technology combinations have achieved comprehensive benefits of increasing yield per mu by 6.3%–15.7% in different regions in demonstrations in Jiangsu, Northeast China, South China, etc.
The implementation of technology is inseparable from policy support and farmers’ participation. Through the “Science and Technology Courtyard” model, researchers have transformed complex technical parameters into simple operation standards such as the “three-looking fertilization method” (looking at seedling condition, soil, and weather), accelerating the large-scale application of green technologies. In the future, with the promotion of these technologies, it is expected to improve the nitrogen use efficiency of rice in China, reduce greenhouse gas emissions from paddy fields, and contribute to ensuring food security and sustainable agricultural development.
Journal
Frontiers of Agricultural Science and Engineering
Method of Research
Experimental study
Subject of Research
Not applicable
Article Title
Integrated innovation and application of green high-yield and high-efficiency technologies of rice in China
How can green technology achieve a win-win for increased food production and environmental protection?
Higher Education Press
image:
Image
view moreCredit: Wen-Feng CONG‡ , Hao YING‡ , Feiyu YING , Zhichao AN , Jianbo SHEN , Fusuo ZHANG
As the most populous country in the world, China feeds 19.1% of the global population with only 8.6% of the world’s arable land. This achievement has been built on a long-standing agricultural model that heavily relies on high fertilizer inputs—China accounts for 32% of global nitrogen fertilizer use, far exceeding that of most countries. However, this “high input, high output” approach has raised concerns: excessive fertilization has led to soil acidification, nitrate pollution in water bodies, PM2.5 emissions, and other environmental issues, which in turn restrict agricultural sustainability. The challenge is how to ensure food security while reducing environmental costs—a common dilemma faced by global agriculture.
Recently, Professor Wenfeng Cong et al. from China Agricultural University proposed a solution called “green technology”, validated through over 12,000 field comparison trials conducted via a nationwide collaborative network. This research not only addresses the aforementioned challenges but also introduces a novel agricultural research paradigm—the “12345” model. This model emphasizes starting from actual production needs and resolving the dual contradictions between high yield and environmental protection, as well as economic growth and ecological preservation, through multidisciplinary collaboration and participation from multiple stakeholders. The relevant paper has been published in Frontiers of Agricultural Science and Engineering (DOI: 10.15302/J-FASE-2025630).
The core of “green technology” is to optimize the “soil-crop-microbe” system to achieve a synergy of “high yield, high efficiency, and low pollution”. Specifically, it includes three key directions. First, constructing high-yield populations by breeding dense-tolerant varieties, adjusting planting densities, or promoting intercropping systems like corn and fava beans to fully utilize light and heat resources. Second, enhancing efficient rhizosphere regulation by using smart fertilizers that precisely match the nutrient needs of crops at different growth stages, or utilizing ammonium nitrogen to promote root growth and phosphorus uptake, thereby improving fertilizer utilization rates. Third, cultivating healthy soils through methods such as combined application of organic and chemical fertilizers and no-till practices to improve soil structure and microbial diversity, providing a foundation for high yields.
What are the effects of this technology? The research team integrated annual field trial data from 12,403 sites conducted between 2005 and 2020 through the national collaborative network. The results showed that compared to conventional farming practices, green technology increased food production by 21%–87% without significantly increasing nitrogen fertilizer inputs, improved nitrogen utilization efficiency by 24%–32%, and reduced nitrogen loss and greenhouse gas emission densities by 50%–56% and 31%–47%, respectively. By 2015, approximately 20.9 million farming households across 452 counties had adopted this technology, covering an arable area of 40 million hectares.
The paper notes that in the context of rising fossil fuel costs, future increases in food production can no longer rely on “piling on” chemical fertilizers; instead, they must achieve “less input, more output, and low pollution” through enhanced efficiency. China’s practices demonstrate that this goal is entirely feasible—if green technology is widely adopted, the impact of Chinese agriculture on global resource consumption, nitrogen and phosphorus loss, and greenhouse gas emissions will be significantly reduced, while also contributing to the achievement of multiple United Nations Sustainable Development Goals.
From theoretical paradigms to farmer practices, this research not only provides a viable pathway for resolving the contradiction between “high yield and environmental protection” but also serves as a reference model for other countries aiming for a green transformation in agriculture through the application of green technology.
Journal
Frontiers of Agricultural Science and Engineering
Method of Research
Experimental study
Subject of Research
Not applicable
Article Title
Green technology for increasing grain crop production and efficiency: innovation and application in China
How to grow more food with fewer resources?
Higher Education Press
image:
Image
view moreCredit: Fulin ZHAO1 , Xingbang WANG1 , Wushuai ZHANG1 , Peng HOU2 , Qingfeng MENG3 , Zhenling CUI4,5 , Xinping CHEN1,4
Global agriculture is facing a dual challenge: ensuring food security for a growing population while reducing the environmental costs associated with production. As a major agricultural country, China has long relied on a resource-intensive model for food production. While this approach has addressed the issue of food sufficiency, it has also led to increased greenhouse gas emissions, soil degradation, and water body eutrophication. Data shows that in 2019, nearly 70% of China's farmland was classified as low to medium productivity. Thus, achieving a green transformation while ensuring food security has become a critical issue in the agricultural sector.
Recently, Associate Professor Wushuang Zhang et al. from Southwest University, China Agricultural University, and the Chinese Academy of Agricultural Sciences systematically reviewed the practices and achievements of green technology innovations in major food crops from 2000 to 2022. They aimed to answer the question: how can China’s agriculture achieve a balance between “high yield” and “high efficiency” amid increasing resource constraints? The related paper has been published in Frontiers of Agricultural Science and Engineering (DOI: 10.15302/J-FASE-2025633).
Over the past 20 years, China’s food production has delivered impressive “green results”. Statistics indicate that, by 2022, the total output of the three major staple crops (rice, wheat, and corn) had increased by 58% compared to 2000, with corn experiencing a remarkable 162% increase. During the same period, the planted area only expanded by 8.6%, with the increase in production primarily driven by improvements in yield per unit area. Specifically, the yield per unit area for wheat rose by 56.7%, corn by 40%, and rice by 12.9%. Even more noteworthy is the improvement in resource efficiency. Fertilizer application peaked in 2016 and decreased by 0.83 million tons by 2022, including a 9.4% reduction in nitrogen fertilizer use. Nitrogen utilization efficiency improved from 27.5% in 2000 to 41.3% in 2022, meaning that more food was produced with less fertilizer.
Behind these changes are a series of breakthroughs in green technologies. For instance, the “Integrated Soil-Crop System Management (ISSM)” technology optimizes variety selection, sowing time, and planting density to enhance both light energy utilization and nutrient supply efficiency. Research shows that after applying this technology in North China, corn yields increased by 91.2% compared to traditional planting methods, while also reducing active nitrogen loss by 30% and greenhouse gas emissions by 11%. Another example is the “Root Zone Nutrient Regulation Technology”, which precisely matches the nitrogen needs of crops at different growth stages, resulting in an 8% increase in corn yield while reducing nitrogen fertilizer use by 25%. The “Rhizosphere Nutrient Regulation Technology” focuses on the smaller-scale root zone environment, optimizing fertilizer application locations and microbial interactions, leading to a 20.2% increase in rice yield and a 20%–30% reduction in nitrogen fertilizer use.
However, challenges remain significant. With population growth and the development of animal husbandry, China’s demand for food, especially corn, is expected to continue rising, with total corn demand projected to increase by 30% by 2050. At the same time, issues of nitrogen and phosphorus surplus in farmland are prominent, and the utilization rate of organic resources remains low, with much potential yet to be unlocked.
To address these challenges, the researchers proposed four major strategies: (1) precision management of organic resources; (2) promotion of enhanced-efficiency fertilizers; (3) promotion and adoption rhizosphere nutrient regulation technology; and (4) new technologies such as intelligent nutrient management.
The researchers also predicts that if “Integrated Soil-Crop System Management” is fully implemented, China’s total output of rice, wheat, and corn could increase by 45.8 million tons, 115 million tons, and 360 million tons, respectively, by 2050, significantly reducing environmental costs while ensuring food security.
Journal
Frontiers of Agricultural Science and Engineering
Method of Research
Experimental study
Subject of Research
Not applicable
Article Title
Innovations in green technology for increasing major grain crop production and efficiency in China
How to achieve “more grain with less pollution”?
image:
Image
view moreCredit: Xiangwen FAN1 , Wenqi MA2 , Zhaohai BAI1 , Fusuo ZHANG3 , Lin MA
As the global population continues to grow and climate change intensifies, the challenges facing agriculture have become increasingly complex. There is a need to meet the rising demand for food while also reducing environmental costs associated with fertilizer overuse, soil degradation, and greenhouse gas emissions. Traditional agricultural research has often focused on single objectives, such as the early “Green Revolution”, which pursued high yields, or later shifts toward organic farming aimed at reducing external inputs. However, it has been difficult to simultaneously satisfy the dual demands of “more grain” and “less pollution”. Is there a method that can enhance agricultural productivity while also ensuring efficient resource use and ecological protection?
Recently, Professor Lin Ma et al. from Nanjing University, China Agricultural University, and Hebei Agricultural University proposed a new agricultural system research method that combines “top-down” and “bottom-up” approaches, providing a viable pathway to address this dilemma. The related paper has been published in Frontiers of Agricultural Science and Engineering (DOI: 10.15302/J-FASE-2025628).
The core of this new method is to construct an agricultural innovation system that balances food security, resource efficiency, and environmental sustainability through interdisciplinary integration and multi-scale collaboration. The “top-down” approach starts with national food security goals, setting a minimum grain production baseline that is then broken down by region. This involves considering local water resources, land carrying capacity, and greenhouse gas emission thresholds to develop specific technical pathways—such as identifying areas that require optimized fertilization or where water-saving technologies should be promoted. Ultimately, these plans are implemented through policy guidance and technical training.
Conversely, the “bottom-up” approach is more closely aligned with frontline production. Researchers embed themselves in rural areas, utilizing the “Technology Backyard” as an innovative platform to collect actual production data from smallholders. They diagnose key bottlenecks that limit yields and develop targeted technologies, such as drought-resistant varieties and precision fertilization techniques. These technologies are then integrated into replicable models for promotion in regions with similar natural conditions.
Notably, the “Technology Backyard” plays a critical role. It serves as both a problem collection station and a testing ground for technologies. Researchers live and work alongside farmers, directly obtaining real data from the planting process. Additionally, the backyard acts as a hub for technology dissemination, validating and improving laboratory results in the field before teaching them to more farmers. For example, in the corn-growing areas of the North China, the researchers identified issues in traditional farming practices, such as irrational nitrogen fertilizer application and sparse planting density. They developed technologies like “Dynamic Nitrogen Supply in Root Zones” and “High-Yield Dense Planting”. After applying these techniques in 66 farmers’ test fields, the average corn yield reached 13 tons per hectare—almost twice that of traditional practices—without increasing nitrogen fertilizer use. Similar cases abound: optimizing the layout of livestock and poultry farming could reduce nitrogen pollution exposure for 90% of the population; adjusting crop planting structures could meet future food demands while reducing active nitrogen loss by 18% and greenhouse gas emissions by 20%.
This method integrates “top-down” systematic planning with “bottom-up” frontline innovation, forming a complete chain from national goals to farmer practices. On one hand, the macro objectives ensure that technical directions do not deviate from the overarching priorities of food security and ecological protection; on the other hand, frontline data make technologies more grounded, addressing the disconnect between past research outcomes and production needs. Currently, relevant technologies have been applied in multiple major agricultural production areas in China, enhancing the planting efficiency of smallholders and providing a “Chinese solution” for global agricultural sustainability. For instance, similar “Technology Backyard” models have begun to be promoted in parts of Africa and Southeast Asia, helping local farmers increase production while reducing environmental burdens.
Journal
Frontiers of Agricultural Science and Engineering
Method of Research
Experimental study
Subject of Research
Not applicable
Article Title
Enhancing green productivity and efficiency through innovative approaches to agricultural system research
Can green technologies resolve the “dilemma” in wheat production?
image:
Image
view moreCredit: Gang HE1,2,3 , Wanyi XIE1 , Lei FAN1 , Xiaotian MI1 , Zhaohui WANG1,2
As the world’s largest wheat producer, China’s annual wheat output reaches 136 million tons, and the stability of its production is directly related to global food security. However, in recent years, China’s wheat imports have continued to rise, reaching 9.96 million tons in 2022. Meanwhile, environmental problems caused by excessive fertilizer application have become increasingly prominent. How to ensure output while reducing resource consumption and environmental costs has become a core issue for sustainable agricultural development.
Recently, a research team led by Professor Zhaohui Wang from the College of Natural Resources and Environment at Northwest A&F University proposed a technical framework for green wheat production and a regionally adapted model, providing ideas to solve this problem. The related paper has been published in Frontiers of Agricultural Science and Engineering (DOI: 10.15302/J-FASE-2025606).
The study constructed a green wheat production framework from three systematic levels: soil, root zone, and canopy. The soil system focuses on improving fertility and stress resistance, and improves soil structure through technologies such as organic fertilizer application and straw returning. The root zone system achieves precise matching of nutrient supply and demand by optimizing water and fertilizer management. The canopy system enhances light energy utilization through variety improvement and planting density regulation. This multi-system collaborative technical system breaks through the limitations of traditional single technologies and lays a foundation for integrated innovation.
Based on this framework, the researchers verified the practical effects of several core technologies. In terms of soil improvement, the combined application of organic fertilizer and mineral fertilizer can increase soil organic carbon sequestration efficiency to 26% and wheat yield by 15.1%; straw returning increases soil carbon storage by 302 kg·ha–1·yr–1 and yield by 6.6%. In the field of nutrient management, the combination of deep fertilizer application and slow/controlled-release fertilizer technology can increase nitrogen use efficiency by 8.3%–16.6% while reducing nitrogen loss by 24%–50%. In terms of water management, drip irrigation technology saves 41% more water than traditional flood irrigation while increasing yield by 5%, and precise regulation of irrigation timing can further increase yield by 7.1%.
Targeting the characteristics of different agricultural ecological zones, the researchers developed differentiated technical models. In the dryland of the Loess Plateau, the “Year-round Plastic Mulching” (YPM) technology increases soil water storage by 7% and yield by 11% through full-period mulching, while reducing nitrate leaching by 63%. In the Guanzhong irrigation area, the “Efficient Nutrient and Water Management” (ENWM) model, through the coupling of soil nitrate monitoring and drip irrigation, reduces irrigation water and nitrogen fertilizer usage by 33% and 30% respectively, while increasing yield by 10% and nitrogen partial factor productivity by 57%.
To promote technology transformation, the research constructed a “Multi-subject Joint Innovation Technology” (MJIT) promotion model. Guided by policies, this model integrates resources from universities, enterprises, agricultural technology extension departments, and other parties. Through the “Science and Technology Courtyard” zero-distance service model, the technology has been applied to over 100 kha of farmland, achieving comprehensive benefits of “yield increase, fertilizer saving, and water saving”.
The study points out that future efforts should focus on strengthening the research and development of regionally adaptable technologies and improving market-oriented promotion mechanisms. This achievement provides a replicable technical path for the green transformation of China's wheat industry and offers a reference for the coordinated development of global food security and ecological protection.
Journal
Frontiers of Agricultural Science and Engineering
Method of Research
Experimental study
Subject of Research
Not applicable
Article Title
Innovation and application of technology models for wheat green production in China
How can science and technology solve the problem of increasing grain yield per unit area?
Higher Education Press
image:
Image
view moreCredit: Peng NING1,2 , Xiaojie FENG1 , Zhanhong HAO1 , Songlin YE2 , Dongyu CAI3 , Kaiye ZHANG1 , Xinsheng NIU2 , Weifeng ZHANG1,2
The North China Plain is an important “granary” in China, with its winter wheat and summer maize planting areas accounting for 73.6% and 30.6% of the national total for wheat and maize respectively. However, its agricultural production has long been trapped in the dilemma of “high input, low efficiency”—fertilizer usage has increased more than 4 times compared with 40 years ago, while grain output has only risen by 1.2 times. Problems such as over-exploitation of water resources and soil degradation have also become increasingly prominent. How to balance agricultural production and ecological protection while ensuring food security?
Recently, a team led by Professor Weifeng Zhang and Peng Ning from the College of Resources and Environmental Sciences at China Agricultural University proposed a sustainable production pathway to achieve an annual yield of 22.5 t·ha–1 in the winter wheat-summer maize rotation system on the North China Plain, providing a scientific reference for solving this problem. The related paper has been published in Frontiers of Agricultural Science and Engineering (DOI: 10.15302/J-FASE-2025618).
The study points out that the current average annual yield of winter wheat and summer maize for farmers on the North China Plain is 12.8 t·ha–1, while the highest yield record in the region has reached 28.1 t·ha–1, indicating huge potential for grain yield increase. However, in traditional planting patterns, excessive fertilizer application leads to low nutrient use efficiency, continuous depletion of regional groundwater, soil organic matter content being only one-third of that in U.S. farmland, and frequent extreme climate events (such as late frost and drought) threatening crop growth. If the existing model remains unchanged, the growth in food demand by 2050 will further exacerbate resource and environmental pressures.
Through the coordinated regulation of multiple factors including “soil-crop-climate-management”, it is possible to reduce resource input while increasing yield. For example, in terms of optimizing the cropping system, appropriately delaying the sowing date of winter wheat and extending the filling period of summer maize can increase the utilization efficiency of light and heat resources at a rate of 71.7 kg·ha–1·yr–1; adopting the “four dense and one sparse” wide-narrow row planting technology combined with shallow-buried drip irrigation can realize precise synchronous management of water and fertilizer, reducing nitrogen input compared with traditional models while increasing wheat and maize yields.
Soil improvement is another key approach. Long-term application of organic fertilizer and straw returning can increase soil organic matter content. When soil organic matter reaches 20–30 g·kg–1, crop yield can increase by about one-fifth, and the soil's ability to retain water and fertilizer can be enhanced. Deep plowing can break the plow layer, improve soil permeability, and when combined with no-tillage technology, it can reduce carbon emissions and achieve “carbon sequestration in soil”.
At present, the aging of agricultural labor force on the North China Plain is intensifying, and traditional experience-based planting is difficult to meet the needs of modernization. Through the “Science and Technology Courtyard” model, researchers reside in villages to conduct experiments together with farmers, transforming complex technologies into “easy to learn and use” localized solutions. For example, in the practice in Quzhou County, Hebei Province, after farmers participated in technology design, wheat and maize yields increased by 7.2% and 11.4%, respectively, and nitrogen use efficiency improved by 27%–28.1%, proving that the “scientist + farmer” collaborative innovation is an effective path for technology promotion.
The study suggests that in the future, efforts should be made to promote sustainable crop production from the following aspects: first, prioritize the strengthening of agricultural infrastructure construction and the improvement of cultivated land soil quality; second, accelerate the breeding of superior crop varieties to fully release high-yield potential; third, promote the interconnection and integration of research methods, focusing on the integrated application of superior varieties, effective methods and advanced technologies; fourth, coordinate national policies and social actions, and strengthen the construction of agricultural technology extension service capabilities.
Journal
Frontiers of Agricultural Science and Engineering
Method of Research
Experimental study
Subject of Research
Not applicable
Article Title
Pathways for sustainable production to approach the potential yield of winter wheat and summer maize on the North China Plain
New few-shot learning model enhances crop disease recognition accuracy
Nanjing Agricultural University The Academy of Science
By integrating a lightweight dilated contextual adapter (DCon-Adapter) and a weight decomposition matrix (WDM), the model learns efficiently from limited samples, achieving 93.53% accuracy in controlled tests and outperforming existing approaches in real-world settings.
Plant disease recognition technologies have advanced rapidly thanks to deep learning and large annotated datasets, but agricultural applications face unique hurdles. Data collection in the field is expensive and time-consuming, and some diseases are so rare that acquiring sufficient samples is nearly impossible. Few-shot learning offers a solution, enabling models to learn from just a few labeled examples per class. Yet, conventional methods often require pretraining on large, domain-specific datasets—a resource rarely available in agriculture. Foundation models, such as CLIP and DINO, have shown strong performance in zero- and few-shot learning, but their generalization to agricultural imagery is limited by domain differences and class imbalances.
A study (DOI: 10.1016/j.plaphe.2025.100024) published in Plant Phenomics on 28 February 2025 by Ruifang Zhai ’s team, Huazhong Agricultural University, improves plant disease recognition accuracy and generalization in data-limited scenarios, offering a practical solution for real-world agricultural diagnostics.
The researchers implemented PlantCaFo, a few-shot plant disease recognition model, by leveraging pretrained backbone networks from foundation models—CLIP (ResNet-50 image encoder and Transformer text encoder), DINO (ResNet-50), and DINO2 (distilled ViT-S/14). Training was conducted with varying sample sizes (1, 2, 4, 8, and 16 shots) using consistent random seeds. Only the cache model, dilated contextual adapter (DCon-Adapter), and weight decomposition matrix (WDM) were trainable, optimizing efficiency. PlantCaFo and its enhanced variant PlantCaFo* (with Mixup and CutMix augmentations) were trained for 40 epochs using AdamW, with evaluation on fixed-size test sets. Experiments on the PlantVillage dataset revealed that while Tip-Adapter-F performed well in ultra-low-shot settings (2–4 shots), PlantCaFo and PlantCaFo* surpassed it in higher-shot scenarios, outperforming CaFo-Base by up to 4.60% and achieving consistent gains on the more challenging Cassava dataset. Confusion matrices confirmed high classification accuracy and minimal misclassifications. Although runtime on PlantVillage doubled relative to CaFo-Base due to larger data handling, accuracy gains of up to 7.74% justified the trade-off. Generalization tests on an out-of-distribution dataset (PDL) showed strong performance on split1 (single-species diseases) but reduced accuracy on split2 (multi-species diseases with complex backgrounds), indicating domain shift challenges. Ablation studies demonstrated that the DCon-Adapter contributed more to performance than the WDM, with their combination yielding further gains, particularly when coupled with data augmentation. Prompt-based experiments confirmed PlantCaFo’s superior text–image understanding even with simple templates. Visualizations using Smooth Grad CAM++ revealed that, compared to CaFo-Base, PlantCaFo more effectively focused on disease-relevant regions while filtering irrelevant features, albeit with slightly less precise localization due to its broader generalization across species. These results highlight PlantCaFo’s capacity to balance accuracy, efficiency, and adaptability for diverse plant disease identification tasks under data-scarce conditions.
PlantCaFo’s ability to accurately recognize plant diseases from minimal data could transform agricultural diagnostics, particularly in resource-limited settings. Farmers, agronomists, and plant health agencies could rapidly deploy AI-based disease detection tools without the prohibitive costs of collecting large training datasets. This efficiency makes the technology suitable for mobile apps, drone-based monitoring systems, and early-warning platforms that help curb disease spread and reduce crop losses.
##
References
DOI
Original Source URL
https://doi.org/10.1016/j.plaphe.2025.100024
Funding information
This work was supported by the National Key Research and Development Program of China (2023YFF1000100).
About Plant Phenomics
Science Partner Journal Plant Phenomics is an online-only Open Access journal published in affiliation with the State Key Laboratory of Crop Genetics & Germplasm Enhancement, Nanjing Agricultural University (NAU) and distributed by the American Association for the Advancement of Science (AAAS). Like all partners participating in the Science Partner Journal program, Plant Phenomics is editorially independent from the Science family of journals. Editorial decisions and scientific activities pursued by the journal's Editorial Board are made independently, based on scientific merit and adhering to the highest standards for accurate and ethical promotion of science. These decisions and activities are in no way influenced by the financial support of NAU, NAU administration, or any other institutions and sponsors. The Editorial Board is solely responsible for all content published in the journal. To learn more about the Science Partner Journal program, visit the SPJ program homepage.
Journal
Plant Phenomics
Method of Research
Experimental study
Subject of Research
Not applicable
Article Title
PlantCaFo: An efficient few-shot plant disease recognition method based on foundation models
Drones and 3D models unlock new genetic insights into wheat plant height
Nanjing Agricultural University The Academy of Science
By capturing multiple height quantiles rather than a single average, the approach reveals subtle variations within crop plots, enabling more accurate genetic mapping. The technique identified 11 stable height-related genetic loci, including two potential novel ones, and yielded validated molecular markers to accelerate marker-assisted selection in wheat breeding programs.
Wheat (Triticum aestivum L.) provides about one-fifth of global caloric intake. Plant height (PH) is a key agronomic trait, influencing both yield potential and lodging resistance. Excessive height can cause plants to topple, while overly short plants may suffer reduced biomass and photosynthetic efficiency. During the “Green Revolution,” dwarfing genes boosted yields worldwide, but modern breeding still seeks optimal PH to balance productivity and resilience. Traditional field measurements—manually gauging a few representative plants—are labor-intensive, prone to human error, and fail to capture within-plot variation. Advances in high-throughput phenotyping, especially 3D canopy modeling, offer a path to more precise and unbiased PH assessments.
A study (DOI: 10.1016/j.plaphe.2025.100017) published in Plant Phenomics on 27 February 2025 by Yuntao Ma’s & Yonggui Xiao’s team, China Agricultural University & Chinese Academy of Agricultural Sciences, demonstrates that low-cost UAV cross-circling oblique imaging enables highly accurate, multi-level 3D measurement of wheat plant height, uncovering novel genetic loci and providing validated molecular markers to accelerate precision breeding.
This study employed low-altitude UAV cross-circling oblique (CCO) imaging to assess wheat plant height (PH) across multiple environments, comparing its performance with conventional nadir imaging. Both methods were conducted at the same flight altitude and overlap settings in one environment, with 24 additional plots included to increase PH variability. CCO imaging captured more complete canopy details, particularly at plot fronts, and produced denser, more accurate point clouds. PH values were extracted from 11 height quantiles, revealing consistently higher correlation coefficients and lower RMSEs for CCO imaging than nadir imaging. The 90% quantile most closely matched field-measured PH (FM-PH), while lower quantiles risked measuring stem rather than canopy height. Detailed CCO point clouds reconstructed canopy structures at the organ scale, clearly depicting spikes, though side view capture was limited when plots were closely spaced. Analysis of RIL populations showed both FM-PH and multi-level 3D-PH followed normal distributions, with strong correlations across quantiles and high broad-sense heritability (0.775–0.959 within environments; 0.975–0.982 across environments). The 90% and 92% quantiles yielded RMSEs below 2 cm in most cases, with a maximum correlation of 0.99 between FM-PH and 3D-PH. QTL mapping across seven environments identified 106 loci for FM-PH and 3D-PH, with 40 shared and 11 stable loci unique to multi-level 3D-PH. Two potential novel loci—QPhzj.caas-3A.2 and QPhzj.caas-7A.1—were converted into KASP markers, validated in natural populations, and linked to significant PH differences under varied irrigation. Candidate gene analysis pinpointed Rht5, a gibberellin-sensitive dwarfing gene on chromosome 3B, and TaGL3-5A, associated with grain length and weight on chromosome 5A, both supported by KASP validation. These results demonstrate that CCO imaging provides a precise, scalable tool for phenotyping PH, enabling more comprehensive genetic analysis than traditional methods.
The integration of UAV CCO imaging and multi-level 3D-PH analysis provides breeders with a low-cost, scalable, and precise tool for phenotyping plant height. By uncovering genetic loci that conventional measurements might miss, the method enhances the efficiency of marker-assisted selection, speeding the development of high-yield, lodging-resistant wheat varieties. Beyond wheat, this workflow could be adapted for other crops where canopy architecture and height are important, offering new opportunities in precision agriculture and crop improvement.
##
References
DOI
Original Source URL
https://doi.org/10.1016/j.plaphe.2025.100017
Funding information
This work was funded by National Key R&D Program of China (2022ZD0115703), the National Natural Science Foundation of China (32372196, 42271319) and Pinduoduo-China Agricultural University Research Fund (PC2023A02002).
About Plant Phenomics
Science Partner Journal Plant Phenomics is an online-only Open Access journal published in affiliation with the State Key Laboratory of Crop Genetics & Germplasm Enhancement, Nanjing Agricultural University (NAU) and distributed by the American Association for the Advancement of Science (AAAS). Like all partners participating in the Science Partner Journal program, Plant Phenomics is editorially independent from the Science family of journals. Editorial decisions and scientific activities pursued by the journal's Editorial Board are made independently, based on scientific merit and adhering to the highest standards for accurate and ethical promotion of science. These decisions and activities are in no way influenced by the financial support of NAU, NAU administration, or any other institutions and sponsors. The Editorial Board is solely responsible for all content published in the journal. To learn more about the Science Partner Journal program, visit the SPJ program homepage.
Journal
Plant Phenomics
Method of Research
Experimental study
Subject of Research
Not applicable
Article Title
Genetic resolution of multi-level plant height in common wheat using the 3D canopy model from ultra-low altitude unmanned aerial vehicle imager
