With the rapid expansion of wind energy,wind farms are increasingly being developed in clustered configurations.Consequently,it is essential to assess the spatiotemporal evolution of wake effects within large wind far...With the rapid expansion of wind energy,wind farms are increasingly being developed in clustered configurations.Consequently,it is essential to assess the spatiotemporal evolution of wake effects within large wind farm cluster(WFC)and their impact on power generation performance.In this study,the spatiotemporal evolution of wakes and their influence on power generation performance were investigated using a mesoscale meteorological model coupled with a wind farm parameterization scheme.The JiuquanWind Power Base,situated in Gansu Province’s Hexi Corridor,China,was chosen as the study area because of its significant contribution to national wind energy generation.The results indicate that wind profiles exhibit clear diurnal variations:enhanced vertical mixing during the day produces a relatively uniform wind field,while at night the wind field shows strong vertical wind shear with higher wind speeds at hub height.Wake effects reduce near-surface wind speeds,producing wakes that are weaker but vertically extensive during the day,and stronger but more confined at night,thereby moderating ver-tical wind shear near the surface.Under stable conditions,the average wind speed deficit at hub height reaches 33%,with the wake extending up to 450 m;under unstable conditions,the deficit decreases to 16%,and the wake reaches nearly 1000 m.Wake-induced turbulence increases turbulent kinetic energy(TKE)within and above the rotor-swept region,especially at night,while TKE below the rotor decreases due to reduced shear.Inter-farm wake interactions lead to an average capacity factor loss of 56%for the WFC,with the most pronounced losses occurring under low to moderate wind speeds and particularly under stable atmosphere.展开更多
Wind farm operators always need a better maintenance strategy to increase resource utilization efficiency while controlling operation and maintenance costs.However,conventional maintenance decision-making approaches a...Wind farm operators always need a better maintenance strategy to increase resource utilization efficiency while controlling operation and maintenance costs.However,conventional maintenance decision-making approaches are time-consuming and have poor flexibility and adaptability to various scenarios.This study addressed these challenges by using a large language model(LLM)to understand,generate,and plan maintenance strategies for wind farms characterized by various failure modes and maintenance costs.A labelled-data-supervised fine-tuning LLM for maintenance,named LLM4M,is proposed.The proposed LLM4M model is trained on an extensive dataset of mathematical programs for maintenance to generate optimal strategies for wind farms.Compared with other large parameter LLMs,the fine-tuned LLM4M model demonstrates remarkable accuracy,with an error of approximately 2%from the optimal strategy.In addition,the generalization of the proposed LLM4M model has achieved remarkable results.If the LLM4M model correctly generates the maintenance strategy,the maintenance cost deviates from the optimal solution by only approximately 5%.Furthermore,phase transition behavior is observed,which provides considerable guidance for the development of domain-specific LLMs for the maintenance domain.展开更多
With the development of onshore power grid projects in desert,Gobi and desertified regions as well as offshore wind power projects,electric energy requires longdistance transmission due to insufficient local consumpti...With the development of onshore power grid projects in desert,Gobi and desertified regions as well as offshore wind power projects,electric energy requires longdistance transmission due to insufficient local consumption.Modular multilevel converter based high-voltage direct current(MMC-HVDC)has become the mainstream transmission technology for large-scale remote wind farm integration.However,MMC-HVDC electrically decouples wind farms from the main grid,and its interaction with wind farms can trigger various oscillations.In addition,the MMC-HVDC connected wind farm system differs from single-converter grid-connected systems in structure,equipment complexity,and fault evolution,leading to problems such as low inertia,wideband oscillations and poor fault ride-through.To address these issues,coordinated control strategies considering dynamic interactions between wind farms and MMC-HVDC have been widely studied.This paper thus reviews such coordinated control strategies from three aspects:first,introducing common engineering symmetrical monopole/bipolar MMC-HVDC system structures;second,summarizing grid frequency sensing methods and inertia support control strategies for MMC-HVDC and wind farms;third,discussing coordinated oscillation suppression and fault ride-through control strategies,considering direct current(DC)-side faults,alternating current(AC)-side grid faults,and wind farm area faults.Finally,it summarizes deficiencies of existing studies for each challenge and prospects future research directions for MMC-HVDC connected wind farm systems.展开更多
Large-scale offshore wind farm clusters(OWFCs)have been increasingly connected to the power grid,and requires advanced forecasting models to enhance the prediction accuracy of OWFC's power output.This paper propos...Large-scale offshore wind farm clusters(OWFCs)have been increasingly connected to the power grid,and requires advanced forecasting models to enhance the prediction accuracy of OWFC's power output.This paper proposes a multi-source fusion with patch-guided multi-task learning for power prediction of offshore wind farm clusters.Unlike traditional graph-based approaches that rely on predefined topological relationships,which are limited in capturing the highly similar but rapidly changing meteorological conditions among closely spaced offshore farms,the proposed model employs a parameter-sharing multi-task learning network to achieves both independence and correlation among offshore wind farm clusters,followed by utilizing a dynamically weighted multi-task loss function to gradually optimize the network parameters.Moreover,the proposed model applies the patch-guided feature learning module further enables natural alignment and fusion of multi-source data.To demonstrate the performance of the proposed model,experiments were conducted on offshore wind farm clusters in three different regions.The results show that the proposed model can obtain an average accuracy improvement of around 24.31%for MAE and 19.l4%for RMSE,ensuring prediction accuracy and robustness.展开更多
In wind farms,the wake effect from upstream wind turbines(WTs)reduces the efficiency of downstream counterparts and overall power generation.Yaw control of upstream WTs can mitigate wake effects and maximize power out...In wind farms,the wake effect from upstream wind turbines(WTs)reduces the efficiency of downstream counterparts and overall power generation.Yaw control of upstream WTs can mitigate wake effects and maximize power output,but it must also consider structural impacts to prevent excessive fatigue loads.This paper presents an optimization framework for managing both power and fatigue loads in wind farms using machine learning and multi-objective optimization algorithms.The framework aims to maximize power generation while minimizing fatigue loads.Simulations under complex inflow conditions and machine learning methods are used to rapidly predict power and fatigue loads.The non-dominated sorting genetic algorithm III then optimizes the yaw angles of WTs to achieve optimal power output with minimal fatigue loads,enhancing wind farm efficiency over its lifetime.Results show that the proposed strategy increases power by 8.4%compared to no yaw control and reduces fatigue loads by 12.5%and 4.8%under different objectives,with only a 1.5%reduction in power compared to power-maximizing yaw control.This framework can be extended to other multi-objective optimization problems in wind farms.展开更多
The rapid development of wind energy in China since 2000 has raised concerns about its impacts on local climate and vegetation.Despite regional and local studies,a comprehensive national assessment is lacking.Here,we ...The rapid development of wind energy in China since 2000 has raised concerns about its impacts on local climate and vegetation.Despite regional and local studies,a comprehensive national assessment is lacking.Here,we analyzed the effects of 675 onshore wind farms,representing>90,000 identified wind turbines in China,on land surface temperature(LST)and vegetation using Moderate-resolution Imaging Spectroradiometer(MODIS)satellite data from 2003 to 2022.We found a daytime cooling effect of-0.05±0.48℃(mean±STD)and a nighttime warming effect of 0.06±0.28℃across all wind farms.The construction of wind farm infrastructure initially reduced peak normalized difference vegetation index(NDVI)by-0.006±0.036,and this adverse impact weakened over time(-0.004 after 7 years),indicating vegetation recovery.The wind farm impacts varied by land cover type.The nighttime warming was largest for barren lands(0.19℃),followed by croplands(0.10℃),grasslands(0.07℃),and forests(0.01℃).These differences contributed to increasing night warming from southern to northern China.The adverse vegetation impacts were largest for forests(-0.010),followed by grasslands(-0.008)and barren lands(-0.003),with croplands(0.001)being almost unaffected.Correlation analysis identified precipitation and mean LST as significant factors influencing spatial variations in nighttime LST impact,with greater vegetation decline reinforcing night warming.Our large-scale analysis provides comprehensive evidence of the heterogeneous environmental impacts of wind farms across China,informing the sustainable development of wind energy.展开更多
This work presents a comprehensive biochemical landscape of pea seeds captured by untargeted LC-MSbased metabolomics of ten pea cultivars grown at three Danish field sites following different agricultural practices.Mo...This work presents a comprehensive biochemical landscape of pea seeds captured by untargeted LC-MSbased metabolomics of ten pea cultivars grown at three Danish field sites following different agricultural practices.More than 1200 metabolite features were detected in methanolic extracts of pea seed flours.Of these,nearly 300 features were identified using mass spectral libraries and advanced computational tools.Approximately 40 metabolites were found to be associated with location effect,independent of cultivar type.Organically grown pea samples showed lower levels of the main pea triterpene glycoside soyasaponin I and higher levels of nitrogen-abundant amino acids,indicating increased nitrogen availability in soil.More than 100 metabolites were associated with the location-independent cultivar effect.Akooma and Greenway cultivars showed the most distinct metabolome with greater levels of polyunsaturated fatty acids and lipid oxidation products known to give‘beany’off-flavors.The commonly cultivated pea variety,Ingrid,was devoid of compounds derived from the phenylpropanoid pathway including hydroxycinnamic acid amides such as caffeoyl,feruloyl,and coumaroyl aspartates that were present in all other cultivars.Three chloroauxin metabolites,reported here for the first time,were identified through molecular networking within GNPS platform and propagation of annotation from a computationally predicted indole-3-acetic acid catabolite.Overall,the results indicate biochemical adaptation of pea plants to location or agricultural practices as reflected in their seed metabolome.展开更多
This paper develops an operational framework for short-horizon risk management in multistate stochastic systems,with application to wind farm performance.We focus on instantaneous mobility-based indicators derived fro...This paper develops an operational framework for short-horizon risk management in multistate stochastic systems,with application to wind farm performance.We focus on instantaneous mobility-based indicators derived from finite-state continuous-time Markov chains,which capture the local propensity of a system to transition between states.Unlike classical reliability and availability measures,these indicators provide a dynamic description of system behavior.The indicators are interpreted as policy signals to support decision-making under budget constraints.We introduce a state-conditional expected short-horizon loss,representing non-production risk,and use it to evaluate ranking-based intervention strategies.The framework is applied to a global dataset of wind farms.Results show that mobility-based indicators,especially those related to transition intensity,outperform standard availability proxies in identifying high-risk conditions and concentrating expected losses among top-ranked observations.This supports their use as effective tools for data-driven,policy-oriented risk management.展开更多
Porcine reproductive and respiratory syndrome virus(PRRSV)has emerged as a significant threat to the pig farm-ing industry worldwide,resulting in considerable economic losses.However,few reports detail its economic im...Porcine reproductive and respiratory syndrome virus(PRRSV)has emerged as a significant threat to the pig farm-ing industry worldwide,resulting in considerable economic losses.However,few reports detail its economic impact on the pig farming sector.A study was conducted on 23 breeding pig farms in Hubei Province from January 2021 to December 2023,and the PRRSV infection status and associated economic losses were monitored to address this gap.PRRSV antigens and antibodies were identified through enzyme-linked immunosorbent assay(ELISA)and quantitative reverse transcription polymerase chain reaction(qRT-PCR).Additional monthly production data and weaning costs were gathered.The Kruskal-Wallis nonparametric test was used to assess the differences in production efficiency and weaning costs across various PRRSV infection statuses.Dunn’s test was used to compare multiple groups.The parameter distributions of various variables were determined via@RISK(V.8.5.2)software.Models were developed to evaluate the economic impact of PRRSV infection status on breeding pig farms and assess the losses from a PRRSV outbreak in either provisional PRRSV-negative or PRRSV-positive stable farms.A total of 754 months of monitoring was conducted across 23 breeding pig farms,which included 131 months(17.37%)classified as PRRSV provisional negative,298 months(39.52%)as PRRSV-positive stable,and 325 months(43.11%)as PRRSV-positive unstable.The production efficiency and weaning costs were similar between provisional PRRSV-negative farms and PRRSV-positive stable farms,revealing no significant differences.However,these metrics varied significantly compared with those of PRRSV-positive unstable farms.With respect to provisional PRRSV-negative farms,PRRSV-positive stable farms and unstable farms faced additional annual losses of¥3,135.17 and¥4,898.79 per sow,respectively.Compared with PRRSV-positive stable farms,PRRSV-positive unstable farms incurred an extra annual loss of¥1,763.62 per sow.Upon a PRRSV outbreak on a swine farm,followed by a return to preoutbreak conditions,the average economic loss per sow on provisional PRRSV-negative farms is approximately¥3,061.21.Conversely,PRRSV-positive stable farms face an average loss of approximately¥508.42 per sow.This study provides a systematic evaluation of the economic impact of PRRSV on Chinese pig farms,offering data to support the quantitative assessment of economic losses stemming from PRRSV within the domestic pig farming industry.展开更多
With the rapid development of the new energy industry, large-scale construction of wind farms and the unattended operation and maintenance (O&M) model have become trends in the sector. As the core primary equipmen...With the rapid development of the new energy industry, large-scale construction of wind farms and the unattended operation and maintenance (O&M) model have become trends in the sector. As the core primary equipment for power transmission in wind farms, the operating state of main transformers directly determines the power supply reliability and economic benefits of wind farms. Combined with the 330kV main transformer O&M project of a wind farm in Jiuquan, Gansu Province, this paper focuses on the core technical issues of main transformer state early warning in unattended scenarios. By constructing a multi-dimensional monitoring system, optimizing feature extraction methods and improving the early warning model, accurate pre-judgment of potential faults in main transformers is realized. The research results show that the proposed early warning scheme can effectively improve the fault identification accuracy and reduce the false alarm rate and missing alarm rate, providing technical support for unattended O&M of main transformers in wind farms.展开更多
What is special about the sky bridge near Qingjing Farm?Qingjing Farm in Nantou is a popular place to visit.Every year,it has a Sheep Run Festival!Thousands of visitors go each year to cheer on their favorite sheep an...What is special about the sky bridge near Qingjing Farm?Qingjing Farm in Nantou is a popular place to visit.Every year,it has a Sheep Run Festival!Thousands of visitors go each year to cheer on their favorite sheep and take pictures as they run down the mountain roads.People can enjoy many performances,games and activities during the festival,too.展开更多
The high proportion of renewable energy connected to the grid has led to a significant decline in the power system’s inertia, making frequency stability an increasingly pressing issue. In response to the current lack...The high proportion of renewable energy connected to the grid has led to a significant decline in the power system’s inertia, making frequency stability an increasingly pressing issue. In response to the current lack of active frequency support capability in wind farms, this paper investigates grid-forming control strategies and frequency support methods based on virtual synchronous generator (VSG) technology. By elucidating the fundamental differences between grid-following and grid-forming converters, the paper analyzes the operating principles of grid-forming control that simulates the electromechanical transient characteristics of a synchronous generator. It proposes a frequency support strategy that coordinates inertia response with primary frequency regulation and introduces coordinated energy storage control to enhance the support effect. Simulation results based on PSCAD/EMTDC demonstrate that grid-forming VSG control can raise the minimum system frequency from 49.62 Hz (in grid-following mode) to 49.85 Hz, reduce the steady-state deviation by 60%, and extend adaptability to weak grids with short-circuit ratios greater than 1.0. Coordinated energy storage control further reduces the inertia response time to within 30 ms. The study demonstrates that grid-forming VSG control can effectively enhance wind farms’ active support capability for system frequency, providing a feasible technical pathway for frequency stability in power systems with a high proportion of renewable energy.展开更多
At dawn in Wufu Village,in Dinghu District of Zhaoqing City,Guangdong Province,the newly restored Chaoyangli Cultural Retreat is already welcoming its first visitors.A visitor surnamed Chen,who has travelled from Guan...At dawn in Wufu Village,in Dinghu District of Zhaoqing City,Guangdong Province,the newly restored Chaoyangli Cultural Retreat is already welcoming its first visitors.A visitor surnamed Chen,who has travelled from Guangzhou with his child,stops in front of the Chaxi Academy to admire local intangible cultural heritage crafts.展开更多
This paper addresses the complexity of wake control in large-scale wind farms by proposing a partitioning control algorithm utilizing the FLORIDyn(FLOW Redirection and Induction Dynamics)dynamic wake model.First,the i...This paper addresses the complexity of wake control in large-scale wind farms by proposing a partitioning control algorithm utilizing the FLORIDyn(FLOW Redirection and Induction Dynamics)dynamic wake model.First,the impact of wakes on turbine effective wind speed is analyzed,leading to a quantitative method for assessing wake interactions.Based on these interactions,a partitioning method divides the wind farm into smaller,computationally manageable zones.Subsequently,a heuristic control algorithm is developed for yaw optimization within each partition,reducing the overall computational burden associated with multi-turbine optimization.The algorithm’s effectiveness is evaluated through case studies on 11-turbine and 28-turbine wind farms,demonstrating power generation increases of 9.78%and 1.78%,respectively,compared to baseline operation.The primary innovation lies in coupling the higher-fidelity dynamic FLORIDyn wake model with a graph-based partitioning strategy and a computationally efficient heuristic optimization,enabling scalable and accurate yaw control for large wind farms,overcoming limitations associated with simplified models or centralized optimization approaches.展开更多
The increasing conversion of agricultural land to organic farming requires the development of specifically adapted cultivars.So far,in tomato there is lack of research for selection of germplasm suitable for sustainab...The increasing conversion of agricultural land to organic farming requires the development of specifically adapted cultivars.So far,in tomato there is lack of research for selection of germplasm suitable for sustainable agroecosystems.In this study,we investigated the genotypic and environmental factors affecting the variation of plant,fruits,and root traits in 39 tomato genotypes grown under organic farming conditions.Four independent experiments were conducted in Italy and Spain across two consecutive seasons in 2019 and 2020.For all traits,the factorial linear regression model to estimate the main effects of genotype(G),location(L),year of cultivation(Y)and their interactions,revealed highly significant(P<0.001)variations,with the G factor being largely predominant for most traits.The implementation of the“which-won-where”,“mean performance versus stability”and“discriminative vs representativeness”patterns in the GGE(Genotype plus Genotype by Environment interaction)analysis,allowed the identification of superior cultivars with high stability across the testing environments.Genomic characterization with 30890 high quality SNPs from dd RADseq genotyping analysis,revealed that a specific cluster of cherry tomato accessions were low performing in terms of yield and fruit weight,on the contrary,showed a high content of soluble solids,which in agreement with GGE analysis.Results of this study provide a framework for the potential use of this locally adapted tomato germplasm to address the needs of more sustainable agriculture.展开更多
The increasing population and continuous urbanization make food security a key consideration in sustainable development.Efficient farming strategies with low environmental footprints are thus increasingly required to ...The increasing population and continuous urbanization make food security a key consideration in sustainable development.Efficient farming strategies with low environmental footprints are thus increasingly required to meet food demands.This study presents a design for environmentally friendly,economical,and modular vertical farming systems,in which vegetables are cultivated in a carbon dioxide(CO2)-enriched atmosphere enabled by direct air capture(DAC)and subjected to artificial light exposure.We established a vertical farming setup and conducted experiments to identify productive cultivation strategies by regulating lighting,CO2concentration,biochar application,and plant species.Additionally,a self-developed DAC rotary adsorber was utilized to achieve stable and efficient CO2enrichment.Compared with the control group,the fresh weight of the vegetables in the experimental groups increased by up to 57.5%.Furthermore,we performed a comprehensive evaluation of the design and demonstrated that integrating photovoltaic-thermal(PVT)and DAC units increased the system’s net present value(NPV)by 157%compared with a conventional design without these units.Importantly,we found it possible to maintain the low carbon footprint of the system(0.468 kg-CO2equivalent·kg−1(CO2eq·kg−1)-vegetable)in the production process.Parametric studies and an application analysis on a global scale reveal the wide adaptability of this strategy to diverse conditions.These findings,together with the modular characteristics of vertical farming systems,highlight the promising potential of this design to increase food security and foster sustainable agriculture.展开更多
The share of wind and solar energy in global energy mix is rising rapidly.Despite their great potential for reducing carbon emissions,poorly planned wind and solar farms may encroach on socio-ecologically sensitive ar...The share of wind and solar energy in global energy mix is rising rapidly.Despite their great potential for reducing carbon emissions,poorly planned wind and solar farms may encroach on socio-ecologically sensitive areas,threatening biodiversity and Indigenous people's traditional land uses.However,these potential risks associated with wind and solar farm development worldwide are poorly understood.Here,we evaluate the potential biodiversity and Indigenous risks from wind and solar energy development by examining the extent to which global wind and solar farms are situated within or adjacent to socio-ecologically sensitive areas.Our analysis revealed that 13,699 wind and solar farms or 14.4%of the farms'total footprint area are within protected areas,critical habitats,and Indigenous people's lands,occupying a total of 26,840 km2of those socio-ecologically sensitive areas.Wind and solar farms overlap with the distribution ranges of 2,310 threatened amphibians,birds,mammals,and reptiles,accounting for 36.3%of the world's 6,362 threatened vertebrate species.The encroachment of solar and wind farms on sensitive areas mostly occurs in economically developed countries with substantial wind and solar energy facilities,while many developing countries in the tropics tend to have a higher proportion of such farms situated within sensitive areas.Compared to wind farms,solar farms pose greater risks to biodiversity and Indigenous people's lands.These findings provide valuable insights into the socio-ecological risks of wind and solar energy development and highlight the urgent need for strategic planning to mitigate the risks.展开更多
This study uses prefecture-level city statistical data from China from 2000-2022 to measure the supply and demand of livestock manure nitrogen nutrients and calculates farmland livestock carrying capacity using the ni...This study uses prefecture-level city statistical data from China from 2000-2022 to measure the supply and demand of livestock manure nitrogen nutrients and calculates farmland livestock carrying capacity using the nitrogen nutrient balance method.We investigate nitrogen supply and demand and livestock carrying capacity in northeast China by comparing emissions from 2000-2022 over the past several decades.The poultry and livestock industry in northeast China has changed significantly over the past two decades:pigs are now the most bred animal and poultry production has increased dramatically.Regional livestock nitrogen emissions are influenced primarily by the size of the local livestock industry chain.Due to regional differences,each region has unique breeding structures.We also predict the anticipated situation in 2050 using the business-as usual scenario.High-risk livestock carrying capacity areas will be concentrated in the northeast and southeast regions,with significantly increased risk indices,compared with those of 2022,particularly in Shenyang,Fushun,and Tieling.Therefore,promoting farming and breeding,improving livestock manure utilization,and returning manure to nearby farmlands are crucial for meeting agricultural green development goals.展开更多
Agriculture extension and advisory services(AEAS)are integral to smart agricultural systems and play a pivotal role in supporting sustainable agricultural development.The study aimed to assess the role of AEAS in stre...Agriculture extension and advisory services(AEAS)are integral to smart agricultural systems and play a pivotal role in supporting sustainable agricultural development.The study aimed to assess the role of AEAS in strengthening climate-smart coastal farming system to enhance coastal agricultural sustainability.A mixed-methods study was conducted in the southwestern coastal region of Bangladesh in 2023,which involved administering a structured questionnaire and conducing face-to-face interviews with 390 farmers.Perceived role index(PRI)was employed to assess the potential role of AEAS.To forecast the perceived role outcomes,the machine learning model was undertaken by utilizing suitable algorithms.Additionally,feature importance was calculated to underpin the significant factors of perceived role outcomes.The findings showed that coastal farming communities held a comprehensive understanding of the role of AEAS.Key roles included diffusion of agricultural innovations,acting as a bridge between farmers and research organizations,using demonstration techniques to educate farmers,training farmers on food storage,processing,and utilization,and promoting awareness and adoption of best practices.The machine learning model exposed a significant relationship between farmers’socio-economic characteristics and their perception behavior.The results identified that factors like innovativeness,awareness,training exposure,access to AEAS,and access to information significantly influenced how farmers perceived the efficacy of AEAS in promoting a smart coastal farming system.However,farmers confronted multiple constraints in receiving demand-driven services and maintaining coastal farm sustainability.These insights can guide concerned authorities and policy-makers in providing AEAS for the purpose of strengthening climate-smart coastal farming system,particularly with a special focus on capacity building programs and machine learning application.Moreover,the outcomes of this study can assist the authorities of similar coastal systems throughout the world to initiate potential strategies for enhancing region-specific agricultural sustainability.展开更多
About 44%of the world’s cocoa is produced in one single country,Côte d’Ivoire.Providing this important raw material,most Ivorian cocoa farmers live in severe poverty,which,despite a multitude of sector interven...About 44%of the world’s cocoa is produced in one single country,Côte d’Ivoire.Providing this important raw material,most Ivorian cocoa farmers live in severe poverty,which,despite a multitude of sector interventions,is still widespread,affecting social and environmental sustainability in cocoa production.In this context,cocoa farmers are still often treated as a homogeneous group of small-scale producers(mainly males),resulting in interventions being conceptualized as one-size-fits-all approaches and failing to deliver support schemes that take farmers’specific conditions appropriately into account.Applying a broader typology approach that combines farm characteristics with farmers’characteristics,this study aims to delineate Ivorian cocoa farmers and their farms into specific types in order to improve advice for targeted sustainability interventions and living income(LI)potentials.Principal component analysis and hierarchical clustering analysis of a household dataset collected in 2022 in five cocoa-growing regions of Côte d’Ivoire were chosen to identify types of male-headed farms.To assure gender sensitive analysis,a female-headed farm type was created artificially.The specific characteristics of the identified types were captured using descriptive analysis.Descriptive statistics and non-parametric tests were then applied to examine the relationships between these farm types and various outcomes.Additionally,a binary logistic model was used to estimate the probability of these links in relation to variables relevant for achieving a LI.Finally,Spearman non-parametric correlation was used to identify eventual differences in the strength of relationships between key variables per farm type.Three different types of male-headed farms are identified:type 1(the most productive and diversified farms with larger size),type 2(middle-sized farms with strong focus on cash crops),and type 3(small-sized farms with a good level of diversification for self-consumption).The artificially created type 4 represents female-headed farms with the smallest size.On average,none of these farm types achieves a LI.However,type 1 shows the smallest LI gap,while type 4 is by far the worst.Our analyses reveal underlying socio-economic factors systematically disadvantaging female-headed cocoa farms,most notably limited access to land and other material assets.The key contribution of this study lies in the empirical identification of the different characteristics of farms in a given farming system,thereby identifying the need for targeted support interventions.Type-specific recommendations are made,showing pathways to provide tailored programs to farmers of different types in order to reduce their LI gaps.展开更多
基金supported by the National Key Research&Development Program of China(Grant No.2022YFB4202104)the National Natural Science Foundation of China(Grant Nos.U24A20196,52166014 and 12302301)+4 种基金Gansu Provincial Key Research and Development Program(Grant No.25YFGA035)Gansu Provincial Education Department Industrial Support Plan Project(Grant No.2025CYZC-026)Central Government Special Funds for Guiding Local Science and Technology Development(Grant No.25ZYJH001)Offshore Wind Power Joint Fund Project of Guangdong Basic and Applied Basic Research Fund(Grant No.2024A1515240050)Shantou University Scientific Research Initiation Grant(Grant No.NTF24029T).
摘要With the rapid expansion of wind energy,wind farms are increasingly being developed in clustered configurations.Consequently,it is essential to assess the spatiotemporal evolution of wake effects within large wind farm cluster(WFC)and their impact on power generation performance.In this study,the spatiotemporal evolution of wakes and their influence on power generation performance were investigated using a mesoscale meteorological model coupled with a wind farm parameterization scheme.The JiuquanWind Power Base,situated in Gansu Province’s Hexi Corridor,China,was chosen as the study area because of its significant contribution to national wind energy generation.The results indicate that wind profiles exhibit clear diurnal variations:enhanced vertical mixing during the day produces a relatively uniform wind field,while at night the wind field shows strong vertical wind shear with higher wind speeds at hub height.Wake effects reduce near-surface wind speeds,producing wakes that are weaker but vertically extensive during the day,and stronger but more confined at night,thereby moderating ver-tical wind shear near the surface.Under stable conditions,the average wind speed deficit at hub height reaches 33%,with the wake extending up to 450 m;under unstable conditions,the deficit decreases to 16%,and the wake reaches nearly 1000 m.Wake-induced turbulence increases turbulent kinetic energy(TKE)within and above the rotor-swept region,especially at night,while TKE below the rotor decreases due to reduced shear.Inter-farm wake interactions lead to an average capacity factor loss of 56%for the WFC,with the most pronounced losses occurring under low to moderate wind speeds and particularly under stable atmosphere.
基金funded by the National Natural Science Foundation of China(72401097,72301016,and 72571015)the Beijing Nova Program,and the Fundamental Research Funds for the Central Universities.
摘要Wind farm operators always need a better maintenance strategy to increase resource utilization efficiency while controlling operation and maintenance costs.However,conventional maintenance decision-making approaches are time-consuming and have poor flexibility and adaptability to various scenarios.This study addressed these challenges by using a large language model(LLM)to understand,generate,and plan maintenance strategies for wind farms characterized by various failure modes and maintenance costs.A labelled-data-supervised fine-tuning LLM for maintenance,named LLM4M,is proposed.The proposed LLM4M model is trained on an extensive dataset of mathematical programs for maintenance to generate optimal strategies for wind farms.Compared with other large parameter LLMs,the fine-tuned LLM4M model demonstrates remarkable accuracy,with an error of approximately 2%from the optimal strategy.In addition,the generalization of the proposed LLM4M model has achieved remarkable results.If the LLM4M model correctly generates the maintenance strategy,the maintenance cost deviates from the optimal solution by only approximately 5%.Furthermore,phase transition behavior is observed,which provides considerable guidance for the development of domain-specific LLMs for the maintenance domain.
基金supported by Smart Grid-National Science and Technology Major Project(Novel Grid-Integrated Transmission System for Far-Offshore Wind Power,No.2024ZD0801300)。
摘要With the development of onshore power grid projects in desert,Gobi and desertified regions as well as offshore wind power projects,electric energy requires longdistance transmission due to insufficient local consumption.Modular multilevel converter based high-voltage direct current(MMC-HVDC)has become the mainstream transmission technology for large-scale remote wind farm integration.However,MMC-HVDC electrically decouples wind farms from the main grid,and its interaction with wind farms can trigger various oscillations.In addition,the MMC-HVDC connected wind farm system differs from single-converter grid-connected systems in structure,equipment complexity,and fault evolution,leading to problems such as low inertia,wideband oscillations and poor fault ride-through.To address these issues,coordinated control strategies considering dynamic interactions between wind farms and MMC-HVDC have been widely studied.This paper thus reviews such coordinated control strategies from three aspects:first,introducing common engineering symmetrical monopole/bipolar MMC-HVDC system structures;second,summarizing grid frequency sensing methods and inertia support control strategies for MMC-HVDC and wind farms;third,discussing coordinated oscillation suppression and fault ride-through control strategies,considering direct current(DC)-side faults,alternating current(AC)-side grid faults,and wind farm area faults.Finally,it summarizes deficiencies of existing studies for each challenge and prospects future research directions for MMC-HVDC connected wind farm systems.
基金supported by the Science and Technology project of State Grid Jiangsu Electric Power Co.,Ltd.(J2025006).
摘要Large-scale offshore wind farm clusters(OWFCs)have been increasingly connected to the power grid,and requires advanced forecasting models to enhance the prediction accuracy of OWFC's power output.This paper proposes a multi-source fusion with patch-guided multi-task learning for power prediction of offshore wind farm clusters.Unlike traditional graph-based approaches that rely on predefined topological relationships,which are limited in capturing the highly similar but rapidly changing meteorological conditions among closely spaced offshore farms,the proposed model employs a parameter-sharing multi-task learning network to achieves both independence and correlation among offshore wind farm clusters,followed by utilizing a dynamically weighted multi-task loss function to gradually optimize the network parameters.Moreover,the proposed model applies the patch-guided feature learning module further enables natural alignment and fusion of multi-source data.To demonstrate the performance of the proposed model,experiments were conducted on offshore wind farm clusters in three different regions.The results show that the proposed model can obtain an average accuracy improvement of around 24.31%for MAE and 19.l4%for RMSE,ensuring prediction accuracy and robustness.
基金supported by the National Natural Science Foundation of China(Grant Nos.12072105,11932006,and 52308498)Natural Science Foundation of Jiangsu Province,China(Grant No.BK20220976).
摘要In wind farms,the wake effect from upstream wind turbines(WTs)reduces the efficiency of downstream counterparts and overall power generation.Yaw control of upstream WTs can mitigate wake effects and maximize power output,but it must also consider structural impacts to prevent excessive fatigue loads.This paper presents an optimization framework for managing both power and fatigue loads in wind farms using machine learning and multi-objective optimization algorithms.The framework aims to maximize power generation while minimizing fatigue loads.Simulations under complex inflow conditions and machine learning methods are used to rapidly predict power and fatigue loads.The non-dominated sorting genetic algorithm III then optimizes the yaw angles of WTs to achieve optimal power output with minimal fatigue loads,enhancing wind farm efficiency over its lifetime.Results show that the proposed strategy increases power by 8.4%compared to no yaw control and reduces fatigue loads by 12.5%and 4.8%under different objectives,with only a 1.5%reduction in power compared to power-maximizing yaw control.This framework can be extended to other multi-objective optimization problems in wind farms.
基金supported by the National Key R&D Program of China(Grant No.2024YFF0811100)the National Natural Science Foundation of China(Grant No.41901115)the 111 Project of China(Grant No.B23027)。
摘要The rapid development of wind energy in China since 2000 has raised concerns about its impacts on local climate and vegetation.Despite regional and local studies,a comprehensive national assessment is lacking.Here,we analyzed the effects of 675 onshore wind farms,representing>90,000 identified wind turbines in China,on land surface temperature(LST)and vegetation using Moderate-resolution Imaging Spectroradiometer(MODIS)satellite data from 2003 to 2022.We found a daytime cooling effect of-0.05±0.48℃(mean±STD)and a nighttime warming effect of 0.06±0.28℃across all wind farms.The construction of wind farm infrastructure initially reduced peak normalized difference vegetation index(NDVI)by-0.006±0.036,and this adverse impact weakened over time(-0.004 after 7 years),indicating vegetation recovery.The wind farm impacts varied by land cover type.The nighttime warming was largest for barren lands(0.19℃),followed by croplands(0.10℃),grasslands(0.07℃),and forests(0.01℃).These differences contributed to increasing night warming from southern to northern China.The adverse vegetation impacts were largest for forests(-0.010),followed by grasslands(-0.008)and barren lands(-0.003),with croplands(0.001)being almost unaffected.Correlation analysis identified precipitation and mean LST as significant factors influencing spatial variations in nighttime LST impact,with greater vegetation decline reinforcing night warming.Our large-scale analysis provides comprehensive evidence of the heterogeneous environmental impacts of wind farms across China,informing the sustainable development of wind energy.
基金funding from Innovation Fund Denmark via the INNOMISSION 3 Partnership AgriFood-Ture program(1152-00001B)financial support from the Novo Nordisk Foundation for the project“PROFERMENT:solid-state fermentations for protein transformations and palatability of plant-based foods”(NNF21OC0066330)。
摘要This work presents a comprehensive biochemical landscape of pea seeds captured by untargeted LC-MSbased metabolomics of ten pea cultivars grown at three Danish field sites following different agricultural practices.More than 1200 metabolite features were detected in methanolic extracts of pea seed flours.Of these,nearly 300 features were identified using mass spectral libraries and advanced computational tools.Approximately 40 metabolites were found to be associated with location effect,independent of cultivar type.Organically grown pea samples showed lower levels of the main pea triterpene glycoside soyasaponin I and higher levels of nitrogen-abundant amino acids,indicating increased nitrogen availability in soil.More than 100 metabolites were associated with the location-independent cultivar effect.Akooma and Greenway cultivars showed the most distinct metabolome with greater levels of polyunsaturated fatty acids and lipid oxidation products known to give‘beany’off-flavors.The commonly cultivated pea variety,Ingrid,was devoid of compounds derived from the phenylpropanoid pathway including hydroxycinnamic acid amides such as caffeoyl,feruloyl,and coumaroyl aspartates that were present in all other cultivars.Three chloroauxin metabolites,reported here for the first time,were identified through molecular networking within GNPS platform and propagation of annotation from a computationally predicted indole-3-acetic acid catabolite.Overall,the results indicate biochemical adaptation of pea plants to location or agricultural practices as reflected in their seed metabolome.
基金financial support from the European Union-NextGenerationEU program,Missione 4 Componente 1,CUP D53D23006470006,MUR PRIN 2022 n.2022ETEHRM“Stochastic models and techniques for the management of wind farms and power systems”by the Italian Ministero dell’Universitàe della Ricerca.
摘要This paper develops an operational framework for short-horizon risk management in multistate stochastic systems,with application to wind farm performance.We focus on instantaneous mobility-based indicators derived from finite-state continuous-time Markov chains,which capture the local propensity of a system to transition between states.Unlike classical reliability and availability measures,these indicators provide a dynamic description of system behavior.The indicators are interpreted as policy signals to support decision-making under budget constraints.We introduce a state-conditional expected short-horizon loss,representing non-production risk,and use it to evaluate ranking-based intervention strategies.The framework is applied to a global dataset of wind farms.Results show that mobility-based indicators,especially those related to transition intensity,outperform standard availability proxies in identifying high-risk conditions and concentrating expected losses among top-ranked observations.This supports their use as effective tools for data-driven,policy-oriented risk management.
基金supported by the Fundamental Research Funds for the Central Universities in China(Project 2662020DKPY016).
摘要Porcine reproductive and respiratory syndrome virus(PRRSV)has emerged as a significant threat to the pig farm-ing industry worldwide,resulting in considerable economic losses.However,few reports detail its economic impact on the pig farming sector.A study was conducted on 23 breeding pig farms in Hubei Province from January 2021 to December 2023,and the PRRSV infection status and associated economic losses were monitored to address this gap.PRRSV antigens and antibodies were identified through enzyme-linked immunosorbent assay(ELISA)and quantitative reverse transcription polymerase chain reaction(qRT-PCR).Additional monthly production data and weaning costs were gathered.The Kruskal-Wallis nonparametric test was used to assess the differences in production efficiency and weaning costs across various PRRSV infection statuses.Dunn’s test was used to compare multiple groups.The parameter distributions of various variables were determined via@RISK(V.8.5.2)software.Models were developed to evaluate the economic impact of PRRSV infection status on breeding pig farms and assess the losses from a PRRSV outbreak in either provisional PRRSV-negative or PRRSV-positive stable farms.A total of 754 months of monitoring was conducted across 23 breeding pig farms,which included 131 months(17.37%)classified as PRRSV provisional negative,298 months(39.52%)as PRRSV-positive stable,and 325 months(43.11%)as PRRSV-positive unstable.The production efficiency and weaning costs were similar between provisional PRRSV-negative farms and PRRSV-positive stable farms,revealing no significant differences.However,these metrics varied significantly compared with those of PRRSV-positive unstable farms.With respect to provisional PRRSV-negative farms,PRRSV-positive stable farms and unstable farms faced additional annual losses of¥3,135.17 and¥4,898.79 per sow,respectively.Compared with PRRSV-positive stable farms,PRRSV-positive unstable farms incurred an extra annual loss of¥1,763.62 per sow.Upon a PRRSV outbreak on a swine farm,followed by a return to preoutbreak conditions,the average economic loss per sow on provisional PRRSV-negative farms is approximately¥3,061.21.Conversely,PRRSV-positive stable farms face an average loss of approximately¥508.42 per sow.This study provides a systematic evaluation of the economic impact of PRRSV on Chinese pig farms,offering data to support the quantitative assessment of economic losses stemming from PRRSV within the domestic pig farming industry.
摘要With the rapid development of the new energy industry, large-scale construction of wind farms and the unattended operation and maintenance (O&M) model have become trends in the sector. As the core primary equipment for power transmission in wind farms, the operating state of main transformers directly determines the power supply reliability and economic benefits of wind farms. Combined with the 330kV main transformer O&M project of a wind farm in Jiuquan, Gansu Province, this paper focuses on the core technical issues of main transformer state early warning in unattended scenarios. By constructing a multi-dimensional monitoring system, optimizing feature extraction methods and improving the early warning model, accurate pre-judgment of potential faults in main transformers is realized. The research results show that the proposed early warning scheme can effectively improve the fault identification accuracy and reduce the false alarm rate and missing alarm rate, providing technical support for unattended O&M of main transformers in wind farms.
摘要What is special about the sky bridge near Qingjing Farm?Qingjing Farm in Nantou is a popular place to visit.Every year,it has a Sheep Run Festival!Thousands of visitors go each year to cheer on their favorite sheep and take pictures as they run down the mountain roads.People can enjoy many performances,games and activities during the festival,too.
摘要The high proportion of renewable energy connected to the grid has led to a significant decline in the power system’s inertia, making frequency stability an increasingly pressing issue. In response to the current lack of active frequency support capability in wind farms, this paper investigates grid-forming control strategies and frequency support methods based on virtual synchronous generator (VSG) technology. By elucidating the fundamental differences between grid-following and grid-forming converters, the paper analyzes the operating principles of grid-forming control that simulates the electromechanical transient characteristics of a synchronous generator. It proposes a frequency support strategy that coordinates inertia response with primary frequency regulation and introduces coordinated energy storage control to enhance the support effect. Simulation results based on PSCAD/EMTDC demonstrate that grid-forming VSG control can raise the minimum system frequency from 49.62 Hz (in grid-following mode) to 49.85 Hz, reduce the steady-state deviation by 60%, and extend adaptability to weak grids with short-circuit ratios greater than 1.0. Coordinated energy storage control further reduces the inertia response time to within 30 ms. The study demonstrates that grid-forming VSG control can effectively enhance wind farms’ active support capability for system frequency, providing a feasible technical pathway for frequency stability in power systems with a high proportion of renewable energy.
摘要At dawn in Wufu Village,in Dinghu District of Zhaoqing City,Guangdong Province,the newly restored Chaoyangli Cultural Retreat is already welcoming its first visitors.A visitor surnamed Chen,who has travelled from Guangzhou with his child,stops in front of the Chaxi Academy to admire local intangible cultural heritage crafts.
基金supported by the Science and Technology Project of China South Power Grid Co.,Ltd.under Grant No.036000KK52222044(GDKJXM20222430).
摘要This paper addresses the complexity of wake control in large-scale wind farms by proposing a partitioning control algorithm utilizing the FLORIDyn(FLOW Redirection and Induction Dynamics)dynamic wake model.First,the impact of wakes on turbine effective wind speed is analyzed,leading to a quantitative method for assessing wake interactions.Based on these interactions,a partitioning method divides the wind farm into smaller,computationally manageable zones.Subsequently,a heuristic control algorithm is developed for yaw optimization within each partition,reducing the overall computational burden associated with multi-turbine optimization.The algorithm’s effectiveness is evaluated through case studies on 11-turbine and 28-turbine wind farms,demonstrating power generation increases of 9.78%and 1.78%,respectively,compared to baseline operation.The primary innovation lies in coupling the higher-fidelity dynamic FLORIDyn wake model with a graph-based partitioning strategy and a computationally efficient heuristic optimization,enabling scalable and accurate yaw control for large wind farms,overcoming limitations associated with simplified models or centralized optimization approaches.
基金supported by the European Union's Horizon 2020 Research and Innovation Programme under Grant No.774244(Breeding for resilient,efficient and sustainable organic vegetable productionBRESOV)by‘RGV-FAO'project funded by the Italian Ministry of Agriculture,Food Sovereignty and Forests。
摘要The increasing conversion of agricultural land to organic farming requires the development of specifically adapted cultivars.So far,in tomato there is lack of research for selection of germplasm suitable for sustainable agroecosystems.In this study,we investigated the genotypic and environmental factors affecting the variation of plant,fruits,and root traits in 39 tomato genotypes grown under organic farming conditions.Four independent experiments were conducted in Italy and Spain across two consecutive seasons in 2019 and 2020.For all traits,the factorial linear regression model to estimate the main effects of genotype(G),location(L),year of cultivation(Y)and their interactions,revealed highly significant(P<0.001)variations,with the G factor being largely predominant for most traits.The implementation of the“which-won-where”,“mean performance versus stability”and“discriminative vs representativeness”patterns in the GGE(Genotype plus Genotype by Environment interaction)analysis,allowed the identification of superior cultivars with high stability across the testing environments.Genomic characterization with 30890 high quality SNPs from dd RADseq genotyping analysis,revealed that a specific cluster of cherry tomato accessions were low performing in terms of yield and fruit weight,on the contrary,showed a high content of soluble solids,which in agreement with GGE analysis.Results of this study provide a framework for the potential use of this locally adapted tomato germplasm to address the needs of more sustainable agriculture.
基金the National Research Foundation(NRF),Prime Minister’s Office,Singapore,under its Campus for Research Excellence and Technological Enterprise(CREATE)program(A-0001032-01-00)the National Natural Science Foundation of China(52376011).
摘要The increasing population and continuous urbanization make food security a key consideration in sustainable development.Efficient farming strategies with low environmental footprints are thus increasingly required to meet food demands.This study presents a design for environmentally friendly,economical,and modular vertical farming systems,in which vegetables are cultivated in a carbon dioxide(CO2)-enriched atmosphere enabled by direct air capture(DAC)and subjected to artificial light exposure.We established a vertical farming setup and conducted experiments to identify productive cultivation strategies by regulating lighting,CO2concentration,biochar application,and plant species.Additionally,a self-developed DAC rotary adsorber was utilized to achieve stable and efficient CO2enrichment.Compared with the control group,the fresh weight of the vegetables in the experimental groups increased by up to 57.5%.Furthermore,we performed a comprehensive evaluation of the design and demonstrated that integrating photovoltaic-thermal(PVT)and DAC units increased the system’s net present value(NPV)by 157%compared with a conventional design without these units.Importantly,we found it possible to maintain the low carbon footprint of the system(0.468 kg-CO2equivalent·kg−1(CO2eq·kg−1)-vegetable)in the production process.Parametric studies and an application analysis on a global scale reveal the wide adaptability of this strategy to diverse conditions.These findings,together with the modular characteristics of vertical farming systems,highlight the promising potential of this design to increase food security and foster sustainable agriculture.
基金supported by the National Natural Science Foundation of China(Grants No.42471287 and 32201422)the Chinese Academy of Sciences。
摘要The share of wind and solar energy in global energy mix is rising rapidly.Despite their great potential for reducing carbon emissions,poorly planned wind and solar farms may encroach on socio-ecologically sensitive areas,threatening biodiversity and Indigenous people's traditional land uses.However,these potential risks associated with wind and solar farm development worldwide are poorly understood.Here,we evaluate the potential biodiversity and Indigenous risks from wind and solar energy development by examining the extent to which global wind and solar farms are situated within or adjacent to socio-ecologically sensitive areas.Our analysis revealed that 13,699 wind and solar farms or 14.4%of the farms'total footprint area are within protected areas,critical habitats,and Indigenous people's lands,occupying a total of 26,840 km2of those socio-ecologically sensitive areas.Wind and solar farms overlap with the distribution ranges of 2,310 threatened amphibians,birds,mammals,and reptiles,accounting for 36.3%of the world's 6,362 threatened vertebrate species.The encroachment of solar and wind farms on sensitive areas mostly occurs in economically developed countries with substantial wind and solar energy facilities,while many developing countries in the tropics tend to have a higher proportion of such farms situated within sensitive areas.Compared to wind farms,solar farms pose greater risks to biodiversity and Indigenous people's lands.These findings provide valuable insights into the socio-ecological risks of wind and solar energy development and highlight the urgent need for strategic planning to mitigate the risks.
基金supported by the National Natural Science Foundation of Youth Project“Research on the Environmental-economic Synergistic Mechanism and Promotion Strategy of Farming and Breeding Circular Pattern in the Black Soil Region of Northeast China”[Grant No.72303087].
摘要This study uses prefecture-level city statistical data from China from 2000-2022 to measure the supply and demand of livestock manure nitrogen nutrients and calculates farmland livestock carrying capacity using the nitrogen nutrient balance method.We investigate nitrogen supply and demand and livestock carrying capacity in northeast China by comparing emissions from 2000-2022 over the past several decades.The poultry and livestock industry in northeast China has changed significantly over the past two decades:pigs are now the most bred animal and poultry production has increased dramatically.Regional livestock nitrogen emissions are influenced primarily by the size of the local livestock industry chain.Due to regional differences,each region has unique breeding structures.We also predict the anticipated situation in 2050 using the business-as usual scenario.High-risk livestock carrying capacity areas will be concentrated in the northeast and southeast regions,with significantly increased risk indices,compared with those of 2022,particularly in Shenyang,Fushun,and Tieling.Therefore,promoting farming and breeding,improving livestock manure utilization,and returning manure to nearby farmlands are crucial for meeting agricultural green development goals.
基金the Science and Technology Fellowship Trust, Bangladesh
摘要Agriculture extension and advisory services(AEAS)are integral to smart agricultural systems and play a pivotal role in supporting sustainable agricultural development.The study aimed to assess the role of AEAS in strengthening climate-smart coastal farming system to enhance coastal agricultural sustainability.A mixed-methods study was conducted in the southwestern coastal region of Bangladesh in 2023,which involved administering a structured questionnaire and conducing face-to-face interviews with 390 farmers.Perceived role index(PRI)was employed to assess the potential role of AEAS.To forecast the perceived role outcomes,the machine learning model was undertaken by utilizing suitable algorithms.Additionally,feature importance was calculated to underpin the significant factors of perceived role outcomes.The findings showed that coastal farming communities held a comprehensive understanding of the role of AEAS.Key roles included diffusion of agricultural innovations,acting as a bridge between farmers and research organizations,using demonstration techniques to educate farmers,training farmers on food storage,processing,and utilization,and promoting awareness and adoption of best practices.The machine learning model exposed a significant relationship between farmers’socio-economic characteristics and their perception behavior.The results identified that factors like innovativeness,awareness,training exposure,access to AEAS,and access to information significantly influenced how farmers perceived the efficacy of AEAS in promoting a smart coastal farming system.However,farmers confronted multiple constraints in receiving demand-driven services and maintaining coastal farm sustainability.These insights can guide concerned authorities and policy-makers in providing AEAS for the purpose of strengthening climate-smart coastal farming system,particularly with a special focus on capacity building programs and machine learning application.Moreover,the outcomes of this study can assist the authorities of similar coastal systems throughout the world to initiate potential strategies for enhancing region-specific agricultural sustainability.
基金This work was conducted in the frame of the accompanying research on strategies for improving farmer families’incomes and sustainable cocoa production funded by the German Federal Ministry for Economic Cooperation and Development(BMZ).
摘要About 44%of the world’s cocoa is produced in one single country,Côte d’Ivoire.Providing this important raw material,most Ivorian cocoa farmers live in severe poverty,which,despite a multitude of sector interventions,is still widespread,affecting social and environmental sustainability in cocoa production.In this context,cocoa farmers are still often treated as a homogeneous group of small-scale producers(mainly males),resulting in interventions being conceptualized as one-size-fits-all approaches and failing to deliver support schemes that take farmers’specific conditions appropriately into account.Applying a broader typology approach that combines farm characteristics with farmers’characteristics,this study aims to delineate Ivorian cocoa farmers and their farms into specific types in order to improve advice for targeted sustainability interventions and living income(LI)potentials.Principal component analysis and hierarchical clustering analysis of a household dataset collected in 2022 in five cocoa-growing regions of Côte d’Ivoire were chosen to identify types of male-headed farms.To assure gender sensitive analysis,a female-headed farm type was created artificially.The specific characteristics of the identified types were captured using descriptive analysis.Descriptive statistics and non-parametric tests were then applied to examine the relationships between these farm types and various outcomes.Additionally,a binary logistic model was used to estimate the probability of these links in relation to variables relevant for achieving a LI.Finally,Spearman non-parametric correlation was used to identify eventual differences in the strength of relationships between key variables per farm type.Three different types of male-headed farms are identified:type 1(the most productive and diversified farms with larger size),type 2(middle-sized farms with strong focus on cash crops),and type 3(small-sized farms with a good level of diversification for self-consumption).The artificially created type 4 represents female-headed farms with the smallest size.On average,none of these farm types achieves a LI.However,type 1 shows the smallest LI gap,while type 4 is by far the worst.Our analyses reveal underlying socio-economic factors systematically disadvantaging female-headed cocoa farms,most notably limited access to land and other material assets.The key contribution of this study lies in the empirical identification of the different characteristics of farms in a given farming system,thereby identifying the need for targeted support interventions.Type-specific recommendations are made,showing pathways to provide tailored programs to farmers of different types in order to reduce their LI gaps.