Category Archives: Agriculture

Pollination is critically important for sustaining the world’s terrestrial biota

Geomembranes are generally welded in the field.The seam is considered a critical point for possible injuries and future leaks in the barrier system.Rollin et al.reported that the total number of faults in geomembranes occurred in landfills, ponds and basins, and 55% of them were found at the seams.Stark et al.evaluated seams made by industry and in the field of a polyolefin geomembrane for a water reservoir project.According to the authors, the seams made by industry can minimize dirt and moisture in the seam, ambient temperature changes and wind.After good statistical data work, the results of this research showed that the seams made by industry were 9% stronger than field seams for peel strength and 10% stronger for shear strength.Moreover, seams according to industry data showed less variability.Lavoie et al.evaluated two different HDPE geomembrane samples exhumed from a municipal landfill leachate pond and a sewage treatment aeration pond.The authors conducted thermal, physical and mechanical analyses.The thermogravimetric analyses showed an altered decomposition behavior for the landfill leachate pond sample, probably caused by the interaction with the leachate.The sewage pond sample presented low stress cracking resistance and low tensile elongation at break, compatible with the dynamic mechanic analysis , which showed an increase in the stiffness.Ewais and Rowe studied an HDPE geomembrane with 1.0 mm of the thickness of about five years immersed in synthetic leachate at different temperatures and in air and water at 55 ◦C.The authors observed changes in the sample’s stress crack resistance values before chemical degradation and before antioxidant depletion for temperatures lower than 70 ◦C.The major SCR value decrease was noted for the sample immersed in leachate at 55 ◦C, reaching 26% of the SCR value of the virgin sample.

This behavior was attributed to morphological changes during aging that affected the interlamellar connections due to the annealing that increased the strength of the inter-lamellar connections and the proposed chain disentanglement mechanism.HDPE geomembranes are often utilized as a flow barrier in landfills, mining facilities, canals, waste liquid ponds,mobile vertical farm and farm ponds.This research evaluated three exhumed high-density polyethylene geomembrane samples from two different shrimp farming ponds after 8.25 years of field exposure for the first pond and 3.0 years of field exposure for the second pond.Thermoanalytical and physical analyses were used to understand the final conditions of the HDPE geomembrane samples.Analyses such as thickness, density, carbon black content, melt flow index , tensile properties, stress crack resistance, oxidative-induction time , thermogravimetry,differential scanning calorimetry,and dynamic mechanical analysis were performed.Using an HDPE geomembrane as a liner of a shrimp farming pond is limited to some places globally, such as the northeast region of Brazil.This work aims to provide new data on the HDPE geomembrane performance in shrimp farming pond applications.This work evaluated three high-density polyethylene geomembrane samples exhumed from two different shrimp farming ponds in the northeast of Brazil with 0.8 mm of nominal thicknesses.The samples called “CAM” and “CAM1′′ represent the same geomembrane installed in a shrimp farming pond and were collected after 8.25 years of field exposure.The CAM sample was exhumed from the bottom liner and had been in contact with the salinized water.The CAM1 sample was exhumed from the same pond as the CAM sample, but it was taken from the slope liner and had been in contact for 8.25 years with environmental conditions.The third sample was exhumed from another shrimp farming pond after 3.0 years of service.This pond has the particularity of being covered with an agricultural plastic film.Figs.1 and 2 show, respectively, the shrimp farming pond representing the CAM and CAM1 samples and the shrimp farming pond representing the CAM2 sample.Table 1 presents the salinized water parameters used for shrimp cultivation.

However, nesting habitats and foraging resources for pollinating insects have decreased severely due to intensification in agricultural landscapes.Developing effective mitigation tools to counteract pollinator losses is essential for satisfactory pollinator diversity and sustainable agriculture.In Europe, agri-environment schemes are among the most valuable instruments to reduce the pressure of agriculture on the environment.The new Common Agricultural Policy of the EU also aims at increasing their share by targeting one-fourth of the direct payments to subsidise eco-schemes.AES offer two main tool types to support biodiversity and thus pollination, i.e.productive and non-productive schemes.Organic farming is one of the most favoured productive management systems where farming intensity is reduced with a preference for mechanical methods over agrochemicals.In their meta-analysis, Tuck et al.found that organic farming increased overall species richness by 30 %, with plants benefiting the most.Pollinator insects also profit from organic farming.Flower strips are flower-rich non-crop landscape features created at arable field edges.The annual or perennial strips are formed to enhance invertebrate pest control and pollination by providing forage, shelter, and nesting places.As AES schemes differ in concept, they should be evaluated against each other to prioritise them for benefits to biodiversity.Functional trait-based assessments offer an effective tool for studying community functionality, redundancy and stability in agroecosystems.Different management types can support various arthropod taxa by maintaining specific plant functional traits and high functional trait diversity.Plant traits associated with visibility, quantity, and quality of floral resources are the most important traits shaping pollinator communities.For example, plants with large, conspicuous inflorescences attract pollinators efficiently , while floral patterns resulting from UV reflection are also important signals to several major pollinator groups.Nectar is the primary reward for pollination; therefore, spatial and temporal variation in nectar volume influences pollinator behaviour and survival.As land-use significantly affects pollination-related flower traits , their diversity can indicate the quality and quantity of floral resources and can be used to evaluate AES schemes for pollinators.Different field parts with different management regimes can support biodiversity and pollination to varying degrees.The interior of arable fields covers the largest area of agricultural landscapes; therefore, its ecological condition may be decisive for biodiversity at the landscape scale.Intensive management within field interiors favours plants with ruderal strategies , and selects for species with early flowering and selfing.Crop edges are also shown to be keystone transitional zones for pollinators and support increased plant diversity compared to field interiors as they are expected to contain both typical arable weeds and species originating from grassy margins.

Linear non-crop landscape elements such as grassy margins along field edges are less intensively managed than field interiors and also provide important shelter and forage resources for various taxa in agricultural landscapes.The overall performance of AES schemes on pollination should be quantified by assessing all characteristic parts of the agricultural fields.Landscape-scale factors should also be considered beside local factors when studying the effects of AES schemes on pollinators and their services.The effectiveness of AES varies by landscape heterogeneity that has two main components.Compositional heterogeneity describes the relative amount of habitats in the landscape, whereas configurational heterogeneity refers to their spatial arrangement.Most studies assessed the effect of landscape composition on pollinator communities and found that pollinator abundance and richness were higher in landscapes comprising more high-quality habitats.The role of landscape configuration is less studied, though it may also be very important as landscapes with smaller fields have higher structural connectivity and more edges providing more floral resources for pollinators.Given this research gap, here we focus on the role of the less studied configurational heterogeneity by assessing landscape-scale mean arable field size.Our study presents a trait-based approach to estimate the potential of two agri-environmental schemes to support pollinators based on floristic composition.For this, we studied plant traits associated with pollinator attraction and reward accessibility.We calculated community-weighted trait means and functional trait diversity in organic and conventional winter wheat fields complemented with sown flower strips and compared them to conventional fields along a gradient of landscape configurational heterogeneity.We tested whether management intensity and landscape heterogeneity select for specific pollination-related plant traits, and result in differing trait diversity by using literature data on species traits.Our specific hypotheses were the following: Pollination-related plant trait diversity is reduced with increasing mean field size within the landscape.Flower strips and organic farming maintain functionally more diverse plant trait composition than control conventional fields.Pollination-related plant trait diversity decreases from grassy margins to field interiors in both conventional and organic fields.We considered only insect-pollinated plant species that may attract pollinators and provide nectar resources.We assessed the following traits: flower size, flower colour, UV pattern in the flower,vertical farming racks reward quantity, and flowering duration.Trait values were collected from the BiolFlor database and identification books.Flower size was expressed as flower diameter.For zygomorphic flowers, the mean of the longest and shortest diameter was calculated.In the case of compact inflorescences that can be regarded as one functional pollination unit, we used the size of the whole inflorescence.We expressed flower colours on a continuous wavelength scale.White-flowered species got the mean value of the visible light spectrum.The presence of UV pattern in the flower was treated as a binary variable.In the case of reward quantity, we considered only nectar resources because of no available information about pollen resources for the majority of the species.Nectar quantity was provided on an ordinal scale with four categories.

The average flowering duration was coded as a continuous variable.For each transect, we calculated the total number and average percentage cover of insect-pollinated plant species.We quantified transect-level community-weighted means for each trait by averaging trait values weighted by species abundance.We tested the effects of management type , transect position , mean field size on the species richness and cover of insect-pollinated plant species, Rao’s quadratic entropy and the CWM and FD values for each trait by using linear mixed-effects models.We used generalized mixed-effects models for species richness with Poisson error distribution and log link function.For each response variable, we started with a model with all two-way interactions of the fixed effects and then selected the minimum adequate model from all possible nested models with the minimum Akaike Information Criterion value.The effect of years was not assessed per se but was treated as a random effect in the analyses.The field, farmer, and landscape identities were also used as random effects nested within each other and crossed with year.We used log transformation for cover values and square root transformation for CWM and FD values if necessary to fulfil test assumptions.We restricted our assessment on those 151 transects that harboured at least one insect-pollinated plant species for flower colour analyses to have meaningful results.We derived marginal and conditional coefficients of determination of the models to estimate the variance explained by the fixed effects and the entire model including both fixed and random effects using the MuMIn.We conducted multiple GLMM tests, therefore, we controlled for the false discovery rate.Post hoc pairwise comparisons were made with the em means package , and were presented in the figures.All analyses were conducted in the R environment.We found that flower strips and organic farming outperformed conventionally farmed fields, as they had the highest insect pollinated plant species richness and supported higher trait diversity favourable for pollinators.Flower strips were superior with the highest cover of insect-pollinated plants and the highest ratio of flowers showing UV patterns.Although annual flower strips doubled the insect-pollinated plant species richness in the grassy margins, the benefits of flower strips completely diminished in neighbouring field interiors.In contrast, organic field interiors could be considered the most favourable pollinator habitat among the studied field interiors based on plant pollination traits.Thus organic farming can enhance pollinators at a larger spatial scale with more environmentally friendly crop production on the whole area of the arable land.We did not find any effect of configurational heterogeneity expressed as mean arable field size.We showed that organic farming outperformed conventional farming practices regarding insect-pollinated species richness, flower attractiveness, and food resources provided for pollinators.This is primarily due to lower overall farming intensity in organic fields.Even organic field interiors offer abundant and diverse flower resources to pollinators compared to conventional farms.Thus, we can consider organic farming the most beneficial method based on pollinators’ resources.In contrast, intensively managed conventional fields supported the lowest richness of insect-pollinated plant species showing that only a few plant species could establish here randomly.Annual flower strips sustained the most favourable vegetation for pollinators, considering the quantity and quality of flower resources.They had six times higher cover of insect-pollinated plants than the edges of the other two farming methods.

Features that focus on pigs vary from features of live pigs to features of meat

We assessed that the first 50 publications contain sufficient information for answering the research questions and thus were used as exclusion criteria. The primary studies with their authors and publication year are included in the appendix . To get insight into the relevance of the topic for this SLR, the results of the broad and narrow search results are grouped by year of publication and presented in Fig. 1 and Fig. 2, respectively. When looking at the number of primary studies published for the broad search query, it seems that there are downward trends in 2010 and between 2014 and 2018. This is however due to the high number of publications that were found and the exclusion criteria that affected publication in those years disproportionately. Many publications were excluded because they were outside of the 50 most relevant publications. However, there is a consistent and sharp increase in the number of publications since 2018 both for the narrow and broad search queries. To get insight into the focus of the primary studies, we have categorized the studies by topic . The categories clearly indicate that majority of the studies were on pig health. Most studies focussed on the presence, spread, or outbreak of diseases, and few studies focussed on the causes and prevention of boar taint. In the selected primary studies, pig health and the performance of pig production were researched often. The production performance was investigated by analysing various parameters, such as the DNA of pigs,rolling bench fat composition of pig meat, meat quality, and muscle thickness. This shows that the production performance of pig farming was mostly measured using the attributes of the meat measured after the animals are slaughtered. Besides pig health and pig production results, other topics were researched.

Several feeding strategies have been developed and evaluated. Further, research was done to test if the origin of meat could be traced and on-farm environmental influencers, such as disease spread in manure and temperature effects on pigs. The narrow search query focused more on the application of ML and the topics of research are summarised in Table 7. According to the search results, most of the research focussed on performance related issues, such as pig performance, feeding strategy, and genetics. Pig health is the second most researched area, focusing mainly on diseases. The trend of using ML for disease detection through image and video analysis is reflected in the number of primary studies found on this topic. Pig detection and behaviour analysis were also performed using ML. Feeding strategy was analysed using basic statistical analysis and using ML. There were also primary studies that focus on detecting the origin of meat, detecting stress in pigs. A detailed description of the topics of research per primary study is shown in Appendix B. As can be seen in Table 8, a variety of data were used for research in the pork industry. Data were largely collected from body and tissue . The rest of the data reported refers largely to measurements of the external characteristics of the animals and their living environment or feed . There were also primary studies that used data on manure and micro-arrays. The type of data used per primary study is shown in Appendix E. The results of the analysis of data from the second search query focused on the application of ML to pig farming are shown in Table 9. The table shows the widely used features used in research. The features used can be grouped into features on pigs and features on climatic conditions.Pork meat information, images and recordings, growth, occupancy and reproduction, genetics, and pig movement have been used to build and validate models. The features on climatic data consist of data about indoor or outdoor climatic conditions, or both. Data used in those primary studies consist of outside temperature, indoor temperature, and relative humidity. The features used per primary study are shown in Appendix H.

The location of data gathering provides insight into the main sources of research data. We have categorized the locations where data are gathered into groups , which shows that most of the data is gathered at farms and slaughterhouses. The farms were sometimes conventional farms where a specific group of animals was monitored closely; other times, they were experimental farms used for precisely monitoring growing conditions and pig behaviour. Slaughterhouses serve as one of the main data gathering locations. The quality and composition of meat are important indicators of production performance; therefore, data from slaughterhouses where these indicators are routinely measured, are important sources of data. Other locations of data gathering for research include laboratories , externally collected data, previous research, or large database where data are shared. Further, data was collected at distribution centres, urban abattoirs, and there was a case of data from a virtual simulation. The data gathering location per primary study is shown in Appendix C. To get an overview of data analysis methods used in the primary studies, the algorithms or principles used were calculated as shown in Table 11. Least squares regression was used most often, followed by logistic regression. The primary studies found by the broad search query did not often use complex algorithms; instead, they applied variations of the linear regression model. All data analysis methods used are shown in Appendix F. All primary studies found by the second search query used ML algorithms. Random forest algorithm was used most, followed by neural networks and support vector machine, as is shown in Table 12. Compared to the primary studies found by the broad search query, those found by the narrow search query used more complex analysis models.

The algorithms used per publication are shown in Appendix G. Eight studies using models based on random forest and/or neural networks recorded an R-squared accuracy scoring higher than 0.7. The challenges stated include difficulty in scaling into large-scale farms due to differences in the environment between the experimental setup and large scale farm, and other features that influence the performance. Table 13 shows what has been studied, and if the research subjects were pigs, it shows the size of the group of pigs studied. A total of 26 primary studies considered live pigs, from which measurements were made. The number of pigs used for research varies significantly, from 5 to 908,582 pigs. Three primary studies considered farms as a whole, focusing on the growing environment of the pigs. Further, data on meat quality and faecal samples were used by some studies. There were also studies that used externally collected data, virtual experimental data, RNA pools, and liver pasts. Appendix D shows the research group types and sizes per primary study. Much research has been using data on genetics, pig breeds, meat quality after processing, disease recognition, and factors influencing diseases. Research that focused on the prevention of negative aspects of the final product is, for example, research done on boar taint and the attempt to reduce it. Boar taint in meat results in a lower price for the carcass. A variety of challenges have been stated in the studies regarding ML. The largest challenge is making the algorithms perform consistently high for all situations. Many studies reported that it was feasible to define a well performing model for the specific research setup and condition the model is trained and tested on but it was in general not possible to guarantee the same level of performance in practice. Mostly, more data and more features are required in order to improve the overall performance of the models developed.The original aim of this study was to review the literature on the use of advanced data analytics techniques in real-life business scenarios. We tried diverse search strings in order to find studies that are not based on experimental set-up be based on the use of data sources from real-life production chains. But the number of results we obtained was low.

We were unable to find any study that fused data from the whole production chain except one by Ma et al. . Ma et al. developed an intelligent feeding equipment and network service platform,roll bench using sensors, software, and monitoring units. This system helped in optimizing feeding methods and improving pig management. We considered, therefore, two search strings, a broad and a narrow search string: the former focussing on general data analytics and the second on machine learning. In both cases, we used more generic terms and relied on exclusion criteria to select the relevant studies. We consulted five scientific databases in this SLR, namely: Science Direct, Scopus, Web of Science, Springer Link, and Wiley. We did not consult Google Scholar initially which returned 16,300 publications, a vast majority of which were irrelevant, and thus filtering out the relevant publications was infeasible. The exclusion criteria we selected ensured that only publications that are accessible and of high quality were considered. The exclusion criterion on year of publication was based on our initial test search results. While research performed 20 years back is likely outdated, the set year allowed us to get a clear view of the trend in the studies published. This made it clear, for instance, that the number of primary studies increased substantially from 2018 onwards, particularly on the application of machine learning in the pork sector. When analysing the results of the SLR, we grouped the attributes such as features, data types, or algorithms we found. Grouping the results allowed us to derive clear conclusions. The detailed results are however included as appendices to this study. As can be seen in the increasing number of primary studies since 2018, the application of ML will be important in the pork sector in the future. When looking at the algorithms used in this context, studies that used features and data on images and recordings often used neural networks, specifically convolutional neural networks algorithm. For example, Cowton et al. , Mukherjee et al. , and Fernandes et al. focused on pig recognition and deviations in behaviour. Studies that used pork meat information often had random forest as best performing algorithm. The focus of the research using pork meat information was to investigate growth performance or the state of health .

What is striking in our review is that most of the research conducted use only groups of pigs selected for an experiment. Most of the studies are conducted in large pig farms; however, the numbers of animals studied within the farms were often small, indicating the studies were conducted on an experimental basis and not on the entire herd. While the focus on a selected group of pigs helps to accurately track and monitor pigs, it does not reflect the real-life situation. Therefore, it remains unclear whether the solutions found in research could be applied in real-life business cases. Out of the 41 primary studies we considered through the broad search query, only 10 involve more than 500 pigs, and of which only two, performed by Naatjes et al. and Haraldsen et al. , used real-time data from farm management systems. Research based on real-life data is scarce and most studies focussed mainly on isolated and one-off problems. We suggest that future research should consider the complexity encountered in real-life circumstances and the integration of data analytics within the management systems so that the analytical results can be used within routine business processes. The Republic of Botswana has an installed capacity of 550 MW with an estimated population of 2.3 million of inhabitants. It is reported that the country has one of the highest national electrification rates in Africa determined as 60 % with 77 % of the population connected to electricity in urban areas and 37 % connected in the rural areas. According to Statistics Botswana, it is stipulated that the Government has set national electricity access target of 82% by 2016 and 100 % by 2030. With an average energy demand of more than 400 MW, Botswana relies mainly on fossil fuels energy sources such as coal, fuel wood, and petroleum and an imported power from South Africa to meet needs of the population. Additionally, the country is also known for its particularity in renewable energy sources such solar, wind and various forms of bio-energy such as bio-fuels and biomass wastes.

Pairwise analyses and significant differences of farming systems were tested with function lsmeans

The significance of the linear model was tested with type III analysis of variance with Kenward-Roger’s method with function anova.OTU data from the amplicon sequencing was normalized using the geometric mean of pairwise ratios method. We performed permutational multivariate analysis of variance using distance matrices with function adonis from vegan 2.5–5 to test the effect of farming system and crop type , and the effect of manure addition, including plants grown in 2018, on microbial community composition. We also conducted 2-D nonmetric multidimensional scaling with stable solution from random starts, axis scaling and species scores with function meta MDS from vegan using the Bray-Curtis dissimilarity index and plotted the NMDS with fitted environmental variables from function envfit in vegan. Fungal and bacterial OTUs indicative for specific rotations were obtained by differential abundance analysis which identified significant groups. Due to the PERMANOVA results we also did additional differential abundance analyses to obtain OTUs indicative for rotation plots that have either received manure or not. The results are presented as paired comparisons between the farming systems for forage and cereal crop rotations, and plots with manure or not, and for the spring and autumn data separately.Bacterial richness, assessed by OTU reads, was significantly higher in the organic rotations compared to the conventional rotations in the autumn data . Furthermore, fungal richness was higher in the organic systems in the autumn data but only significantly for the OMilk rotation. In the spring data bacterial and fungal richness did not differ between farming systems. Neither did bacterial and archaeal 16S rRNA gene copy numbers differ between farming systems in either the spring or autumn data,plastic pots 30 liters whereas fungal ITS copy numbers were higher in the OCer rotation compared to the CCer rotation in the autumn data . The NMDS ordination showed minor differences in bacterial community composition between the farming systems for both crop rotations.

Differences in bacterial community composition, however, were more distinct in the autumn data and higher basal respiration, microbial biomass C and N, as well as extractable C and N variables fitted best with the CMilk rotation. Whereas changes in the fungal community composition between farming systems were more pronounced for both crop rotations and between the seasons . The fungal community composition in the organic rotations were more similar than between the conventional rotations. In the springdata, higher basal respiration rates and microbial biomass C and N variables fitted best with the OMilk rotation. In autumn data, microbial C and N, and extractable C fitted best with the OMilk rotation, and soil pH and basal respiration rates with the OCer rotation, and extractable N with the CMilk rotation. The first PERMANOVA indicated that both farming system and crop type significantly affected the variation in microbial community composition both in the spring and autumn data. However, crop type did not explain the variation in the bacterial community composition in the spring data . Farming system explained 10 and 14% of the variation in bacterial community composition in the spring and autumn data, respectively . Crop type explained 10% of the variation in the autumn bacterial data. In turn, the farming system explained 11 and 14% of the variation in fungal community composition in the spring and autumn data, respectively. Crop type explained even more, 21 and 36% of the variation, in the spring and autumn data, respectively. We also tested the effect of manure in an additional PERMANOVA test, which showed that in addition to farming system manure did not affect the variation in bacterial community composition . However, manure explained additional 14% and 11% of the variation in fungal community composition in the spring and autumn data, respectively. In addition, the impact of manure could not be separated from the impact of crop plant which was grown in all manure fertilized rotation plots in 2018. In the ITS2 based fungal data we could identify on average from 9 to 15 Glomeromycotan AMF OTUs in the spring, and from 12 to 25 in the autumn. In the spring data, the OMilk rotation had the lowest OTU number and the OCer rotation the highest . In the autumn data, the CMilk rotation had the lowest and the OCer rotation the highest AMF OTU number.

The majority of AMF OTUs were shared between the farming systems. PERMANOVA showed that season explained the most variation in the AMF community composition, while crop type and farming system also had an effect. There were large differences in the relative abundances of AMF families between the spring and autumn data of the organic systems . Claroideoglomeraceae was the most abundant family, especially in the autumn data, whereas Acaulosporaceae and Pacisporaceae were not present at all in the CMilk rotation. Archaeosporaceae were abundant in the organic systems and almost absent from the conventional systems. It also seems that Pacisporaceae, while not abundant, is characteristic to the cereal rotations, regardless of the season. The 18S rDNA-based AMF data from composite samples obtained in the autumn identified 38 OTUs with unique genbank accession numbers . Moreover, two Scutellospora species were obtained from the Gigasporaceae family that were not seen in the ITS2 data . Paraglomeraceae, Claroideoglomeracea and Diversisporaceae were the most abundant AMF families based on the 18S data. Organic systems had higher AMF richness; AMF OTU numbers varied from 22 in soil with conventional systems to 35 in soil from OMilk. One out of two Acaulospora sp. and four out of seven Glomus sp. OTUs were only present in the organic systems, with two of them only in the OMilk rotation. Altogether, 19 OTUs were shared between all four rotations. Glomus species were abundant in the OMilk rotation, and similarly to the ITS2 based data, Pacisporaceae was not present in the CMilk rotation, but we could find one Acaulosporaceae OTU present in all rotations, contrary to the ITS2 data. Cereal systems were characteristically associated with Archaeospora trappei, Archaeospora sp, Glomus mosseae, and Pacispora sp. . The more frequent tillage in the conventional system for cereal crops explains partly the lower microbial activity rates and biomasses observed in spring and autumn, a phenomenon which has been observed also by others. Furthermore, the conventional cereal system had bare soil over the winter and is the only treatment without manure addition in the rotation. Thus, the absence of continuous plant cover may have further reduced the microbial biomass C in the spring compared to the respective organic system as earlier reported.

Moreover, the second cut of grass and clover ley was left on the soil surface as a green manure in the organic system for cereal rotation and this may have led to the higher microbial activity and biomass in the following autumn. The lack of chemical agents may also have been reflected in the results since earlier studies have shown negative effects of agrochemicals on soil microbial communities.Higher springtime microbial activity and biomass in the conventional forage crop rotation compared to that of organic cannot be easily explained by the differences in the rotation types as the rotations include various management practices. However, there was 30–40% higher soil P concentration in the conventional system for both cereal and forage crop rotations compared to the organic ones. Indeed, lower water soluble and inorganic P amounts have been reported from organic systems compared to systems receiving synthetic fertilization in a long-term field experiment. Possibly the springtime bacterial community in the conventional forage crop rotation gained competitive advantage from the higher availability of P, since P has been reported to limit bacterial growth in agricultural soils. Later in autumn, the summertime amendments of cow manure would equalize the P availability, and differences between organic and conventional systems for forage crop rotation are no longer detected. The slightly higher autumn pH in the organic system for cereal crop rotation may result from microbial decomposition of fresh plant residues with high N content and mineralization of ammonium which temporarily is known to increase pH. Furthermore, it has been reported that long-term application of manure maintains the soil pH,round plastic pots but inorganic fertilizer decreased it. Consequently, since bacterial growth is known to increase multi-fold with increasing pH, this may explain the increased microbial biomass in the organic system for cereal crop rotation. Alternatively, the increased fungal abundance in the organic system for cereal crop rotation in the autumn indicated that fungi could also be responsible for the higher respiration activity and sequestration of C into their biomass.

Moreover, the organic system for cereal crop rotation had timothy and clover ley as the main crop plant in the sampling year instead of barley as in the respective conventional system. Indeed, higher microbial biomass in production systems including ley grasses have been detected compared to single crop systems only. Our results showed that the farming system induced a clear shift in microbial community composition and that the overall impact of the farming system was about the same magnitude for both bacterial and fungal community composition. Our results are comparable to previous findings that about 10% of variation in microbial communities was explained by the farming practices of conventional and organic systems. Yet, crop type affected fungal community composition in particular, especially in the autumn. A simple explanation would be that changing cultivated plants from barley in the year 2017 to ley in the sampling year 2018 induced a shift in fungal community composition. Furthermore, it is likely that the summertime amendments of synthetic fertilization in the conventional systems have also contributed to the lowered bacterial and fungal richness in the autumn, since the quality of fertilizer is known to impact largely on microbial communities. Thus, the differences in crop rotation, tillage and fertilization practices may all have contributed to the differences in the microbial community but with the current experimental layout we are not able to determine which practices have the strongest impact. Nevertheless, comparisons within the cereal rotations were valuable for indicating the long-term impacts of manure addition and overwintering as bare soil, while comparing the forage crop rotations it was possible to assess the other effects of organic practises beyond manure addition and undersown ley. Moreover, the differences in AMF richness between the farming systems were only moderate and non-existing under the cereal crop rotation. A higher diversity of Acaulospora species, typical to organic systems was also supported by our study. We did not observe Clareideoglomus species to be characteristic to organic systems, instead finding the opposite, which contradicts previous results. Thus, our study supports previous observations which concluded that mycorrhizal diversity is not influenced by the farming system but rather cultivation practices and conditions, and that finding a universal AMF indicator for farming systems is not feasible. Results suggest that the cow manure applied in the forage crop rotation under both the conventional and the organic systems over the years has shaped the bacterial and fungal communities more than any other farming system specific practice, since only a few representatives were typical of either farming system.

This also highlights the commonness of pathogenic fungi in the fields cultivated for fodder and fertilized with manure irrespective of the farming system. On the contrary, AMF richness in the forage crop rotation varied clearly due to the farming system; for instance, Glomus species were indicative in the organic system, while Pacispora sp. was totally missing from the conventional system. Since plants acquire P directly and through their symbiotic AMF, the 40% higher levels of P in the conventional forage crop rotation compared to the respective organic rotation may partly explain the lower AMF richness. However, bacterial representatives were less diverse in the conventional farming system of the cereal rotation compared to the respective organic system. Representative taxa typical of the conventional cereal crop rotation in autumn were affiliated to decomposer and plant-growth promoting and cellulose decomposing bacterial taxa. In contrast, both the spring and autumn data obtained from the organic system for cereal crop rotation revealed a variety of specific taxa with diverse functional roles benefiting soil health. Most of these taxa were also linked to manure fertilization, and included for instance, plant growth-promoting rhizobacteria that include species capable of N-fixation, P solubilization, phytohormone production, and repression of soil-borne plant pathogens, and genera with anti-fungal and antibiotic capability. Furthermore, the spring and autumn data of the organic system for cereal crop rotation contained also season specific taxa, which shared bacterial representatives with similar functional roles.

Organic fertiliser application was slightly increased in KE-C and GH-NC while it was reduced in KE-NC2

In the non-certified case studies, rather small percentages of farms fully followed the rules of OA. In KE-NC1, the case study with the most promising approach, up to 42% of farmers worked organically, without any certification, while in GHNC and KE-NC2 only 16% and 28% did, respectively. Remarkably, in four case studies, the share of smallholder farmers in the control group that did not use conventional inputs was below or around 10%. In KE-NC1, the rate was about 21% . This is contrary to literature, where organic-by-default is often indicated to be common among smallholder farmers in SSA . This indicates the increasing availability and usage of conventional inputs as found by De Bon et al. and Andersson and Isgren . Especially pesticide use among smallholders is far more widespread than commonly assumed . Getting accustomed to conventional input use may lead to a decreased willingness of smallholder farmers to convert to an entirely organic system and agroecological principles . However, input use in SSA varies substantially between countries , therefore these results cannot be extrapolated to other countries. Farmers replacing conventional inputs with either preventive or curative agroecological practices can address nutrient and pest management issues under organic management. These practices include, among others, applications of botanical pesticides, such as neem; preventive practices can involve a more diverse rotation, agroforestry or intercropping systems.We specifically analysed whether the interventions led to an increased uptake of a) practices substituting conventional inputs for pest, disease and weed management, b) practices for substituting mineral fertilisers, and c) further agroecological practices.

Looking at the effect of the interventions on the uptake of AOM practices, our data shows no widespread systematic adoption in any of the case studies.Among the agroecological practices,hydroponic barley fodder system only the diversity of crop rotations was affected positively in GH-NC and KE-NC1 while it was even less diverse in KE-C and KE-NC2. For the remaining components of AOM, we did not find consistent differences between the intervention and control groups. While at least one of the interventions led farmers to adopt POM, AOM was not adopted widely in any of the case studies, despite that all of the interventions aimed at such an adoption. This shows the importance of considering innovation dynamics and transition time frames when introducing organic agriculture to smallholder farmers, as their decision-making is dynamic, multi-dimensional and contextual . Transferring information and skills to famers via group trainings is an important component of capacity development, but needs to be embedded in a long-term process and governance structure, which allows a group of smallholder farmers to learn and explore practices on their own farms and identify ways of combining practices that fit into their specific production system. Compared to applying mineral fertilisers, herbicides and pesticides, agroecological practices are usually knowledge-intensive and require understanding of complex ecological principles . In order to understand the low uptake rates of organic farming practices by farmers in the organic intervention groups, we analysed a) motivations to convert to organic agriculture and b) the implementation challenges as perceived by the farmers. In the two certified organic case studies , high potential economic returns motivated farmers to practice organic farming , while non-financial reasons were less apparent. In the non-certified case studies, the primary motivation to practice organic farming was non-financial with the exception of GH-NC, where 54% of the responding farmers had primarily financial reasons. The differences in motivations and expectations are partly driven by the implementation approach of the intervention. For instance, in KE-NC1, much time was prior invested to make farmers aware of the non-financial benefits of organic farming such as human and environmental health.

While in most case studies, little difference between the responses of adopters and non-adopters could be observed, farmers adopting organic management practices in KE-NC2 and GH-NC had a higher share of financial motivations . The most prominent challenges that the organic intervention farmers faced were: pest and disease damage during crop cultivation and post harvest stages , lack of stable markets , inadequate training and extension services , unavailability of inputs and additional labour required due to weeding . The importance of the challenges was perceived differently in the various case studies. Generally, the Ghanaian farmers perceived the agronomic challenges as more important than the Kenyan ones. Furthermore, our assessment of uptake of farmers is reflected in severity of the challenges, as farmers in KE-C and KE-NC1 who were exposed to these interventions perceive the challenges as overall less severe . While there is only little empirical evidence reported in literature about motivations and challenges of organic farmers in SSA , the technical challenges, such as weed infestation and damage by pests and diseases, are similar to those found in Switzerland by Home et al. although the Swiss farmers in their study reported that these barriers were less severe than they had estimated before conversion to organic. Following the analysis of the effects of the interventions on the adoption of practices, we analysed how OA, as a production system, performed. For this, we compared all farms in each case study, those who worked organically with those who did not, regardless of whether they were part of the intervention group or not . We analysed the differences in yields, inputs, labour and gross margins of the four most widely grown crops in each of the five case studies, using an entropy balancing approach for estimating a sound counterfactual . Among the total of 20 crops analysed from the five case studies, we found four organically managed crops with significantly higher and four crops with significantly lower yields . Input cost was significantly higher for three crops each, while inputs were significantly lower for eight crops and labour was lower for six crops . This resulted in higher gross margins for four organically managed crops, while only one crop had significantly lower gross margins under organic management . Comparing the two certified case studies, farmers practicing OA performed very differently in their productivity and profitability.

Except for reduced cocoa input costs , no significant differences in yields and gross margins between organically and conventionally grown crops could be observed in GH-C. Contrary, we observed higher yields for the economically most relevant crops , while banana yields were lower in KE-C. Input cost was reduced or stayed similar, while labour cost was increased for coffee and macadamia nut in KE-C. Despite labour cost was higher, the gross margins of coffee and macadamia nut increased by 336% and 185%, respectively. Less pronounced differences were found in the non-certified case studies: in GH-NC, yields of organic farms were similar to conventional farms for the four crops, almost no purchased inputs were used and labour was reduced for maize , groundnuts and millet . Gross margins of organically managed crops were at similar levels to their conventional counterparts, except in the case of maize . In KE-NC1, however, organically managed brassica and maize yields were lower than for conventional farmers but other crops were not significantly affected. Maize input cost was significantly lower while labour costs were lower for beans, maize and roots. Overall, no significant differences in gross margins were observed in KE-NC1. On the other hand, in KE-NC2, pea yields were higher, while the other crops remained unaffected by organic management. In this case study, organic farmers significantly reduced input cost for mango while labour cost for peas was higher. In terms of gross margins, there were no significant differences, except for peas . Except for macadamia nut in KE-C,livestock fodder system the organically grown crops in our case studies were not sold with/did not generate a price premium. To assess the importance of local and international markets for organic produce, we therefore tested the impacts of a general 20% price premium for the organic farmers for sensitivity analysis . Assuming an organic price premium, POM gross margins would be higher than the conventional counterpart for bananas in GH-C, coffee in KE-C, beans and maize in GH-NC, roots in KE-NC1, and beans and peas in the case of KE-NC2. The high variability of organic yields and gross margins through organic farming is mostly consistent with findings of meta-studies that are mainly based on data from high-income countries . The methodological difficulties of comparing organic smallholder producers in low-income countries and the resulting uncertainty resulting from the definition of a sound counterfactual led even to a higher uncertainty of impacts of organic agriculture on smallholder yield and profits. Some authors identify strong yield increases due to organic agriculture , while others criticise methodological flaws.

We observed positive effects of organic farming practices at farm level productivity in one of the five case studies . Higher farm level gross margins could neither be achieved in the other certified case study nor in the three non-certified case studies . As, at least in GH-C, organic price premiums were originally supposed to be realised, the farm-level gross margins were analysed under the assumption that at least 20% of price premium could be realised due to the organic management. For the pooled sample over all five case studies, OA had significantly positive impacts on gross margins . Under such an assumption, the organic farmers in four of the five case studies would have realised higher gross margins than the farmers who managed their farms conventionally. Besides KE-C, GH-C , GH-NC and KE-NC1 would also have performed better, while for KE-NC1, there was no significant difference observed. For the pooled sample, OA had significantly positive impacts on gross margins . Our results show that there is no one silver bullet for increasing profitability among smallholder farmers . Profitability increasing effects observed over all the five case studies can be associated with labour input in general and specifically with the number of hours spent on pruning as one of the key specific good agricultural practices for the perennial crops such as macadamia nut, mango and cocoa. When used, the application of organic fertilisers had significantly positive impacts, while conventional fertilisers and pesticides affected the revenues rather negatively. Organic insecticides did not have significant yield effects while copper did. Contrary to findings from field trials , further organic and agroecological management practices resulted mostly in no remarkable economic benefits assessed for the farmers in our study . This could signify that the levels of inputs and practices applied by farmers in our study were still low and not optimal as supported by on-station long-term trial findings in Kenya . Contrary to results from field studies and meta-studies, which report the productivity of organic agriculture crop-specifically , our study shows that contextual factors such as the governance and capacities of smallholder cooperatives are important factors determining the agronomic and economic performance of OA, too.

Many authors suggest that capacity development measures, which are implemented alongside organic projects, are responsible for a large share of the revenue increases that were observed in other studies . In our study, we therefore, controlled for the number of training events from both government and non-governmental organisers. Overall, organic farms had similar numbers of training events and extension visits by NGO and government agents in all case studies, except in KE-C, where the number of governmental and total trainings was even lower than for conventional farms . Government trainings were generally rated lower by organic farmers than by conventional farmers in KE-NC1 and KE-C. On average, NGO trainings were rated better in terms of effectiveness compared to the governmental-based trainings. This indicates a potential for improvement and strengthening of trainings and extension services offered by governmental agents. This further indicates that there are large differences in the perceived quality of NGO training provided to the farmers. There is a great diversity of smallholder farmers in Africa. Much of the on-going controversial discussion about OA is due to a lack of a clear classification and the very heterogeneous characteristics and performances that one can realise on farms that may all be called “organic” on a superficial level . Therefore, based on the results from our study, we propose a terminology for organic farms that can bring more transparency in the debate and can be used to assess the current situation and design tailored public and private policy interventions. Fig. 4 distinguishes organic farming systems according to a) the degree they follow the principles of OA  and b) the intention to work organically .

Scenarios are widely used in research to simulate decision-making environments

Their study found a significant number of behavioural bio-security policies in seven European countries, many of which appeared to match theoretical behavioural change frameworks. However, the most frequent strategies relied on the most basic interventions , and there was little evidence of the systematic use of methods from the behavioural sciences to develop these policies. If this suggests there remains some way to travel before the social sciences are integrated within bio-security policy making, other research continues to highlight the potential value of these approaches. For example, research on the role of information cues reveals that bio-security behaviours can be improved when messages are shown graphically, rather than linguistically or numerically.Drawing on Kahneman and Tversky’s ‘prospect theory’ in which avoiding losses are preferable to accruing gains, Hansson and Lagerkvist show how farmers’ disease management decisions reflect farmers individual assessments of risk. However, when farmers are faced with managing an ongoing disease outbreak, decisions reflect a preference of avoiding losses; gains are only preferred when they seek to prevent future disease outbreaks. Other research has sought to examine how social information and the behaviour of other farmers can influence farmers’ bio-security decisions. Using an experimental simulation, Merrill et al. for instance show that willingness to invest in bio-security decreases when information on environmental disease prevalence is uncertain,stacking pots reflecting an optimism bias that farmers’ herds will not become infected.

Alternatively, when more information is provided about bio-security practices on neighbouring farms, bio-security investment decreases. This work is interesting in that it suggests that social norms of what constitutes ‘bio-security citizenship’ , appropriate conduct or what has been referred to as ‘good farming’ may not be influential in bio-security decision making. Burton suggests that ‘good farming’ refers not only to economic forms of capital, but symbolic cultural capital: the visible demonstration of practical knowledge such as good stockman ship, symbols of appropriate farm maintenance such as clean farmyards and tidy hedgerows, and attributes such as hard work. These symbols are encoded and disseminated within discursive scripts, reinforcing their cultural legitimacy . In this way, good farming acts as a heuristic to provide a strategy to guide, interpret and make decisions in conditions of uncertainty. Other strategies of decision-making are available to farmers, however, and the selection of good farming to guide decisions represents what Sunstein and Ullmann-Margalit refer to as a second-order decision. For Burton and Paragahawewa , the value of the good farmer approach lies in recognising and utilising cultural capital to create more culturally salient agricultural policy. Rather than simply rely on financial payments, they instead recommend the development and incorporation of measures of cultural capital into agricultural policy, and/or restructuring agricultural policy to directly encourage the generation of cultural capital. This may include directly measuring farmers’ ‘skills’ in order to allow them to publicly demonstrate what is valued by the farming community. Whilst Burton and Paragahawewa note that some cultural values might be hard to measure , objectifying cattle purchasing skills may provide a relatively easy way of incorporating the cultural capital of good farming into animal disease management policy. For example, recent research has established a link between farmers’ understandings of good farming and bio-security practices .

In particular, cattle purchasing is likely to be connected to and reflect good farming in a number of ways. Firstly, purchasing cattle risks the introduction and transmission of new diseases to animals within the herd and, for some diseases that can be subsequently transmitted within the local environment, to animals on neighbouring farms. For those farms that need to replace stock, however, different forms of institutional capital – such as certification and ranking schemes – can help provide assurance to the purchaser that they are buying from a good farmer and are running the risk of being labelled a bad farmer by introducing disease into their herd or area. For example, Enticott et al. describe how the number of years a farm has been free from disease effectively establishes a good farming rating that may incentivise improved bio-security when it is required to be displayed at the point of sale. The extent to which these forms of information are a reliable guide to whether the farmer is a ‘good farmer’ may, however, be compromised by farmers’ own spatial understanding of disease transmission and by blaming disease outbreaks on perceived government failings, rather than ‘bad farming’ . Secondly, the avoidance of disease through careful cattle purchasing should allow farmers to display other forms of symbolic cultural capital. An outbreak of bTB, for instance, would lead to a farm’s business being subject to a range of trading restrictions, denying the opportunity to farm with autonomy, which is highly valued by farmers in the farming script of ‘being my own boss’ which symbolises farmers’ success at running their own farm well rather than being told how to farm by government. Indeed, an outbreak of bTB would mean that many farming decisions would be subject to bureaucratic procedures and determined by government officials: farmers would be unable to attend market to sell their cattle. As a result, farms may become over-stocked, and cattle suffer poor welfare. Failing to avoid disease through responsible cattle purchasing therefore compromises farmers’ abilities to display the embodied and practical skills of the good farmer symbolised by good-looking cattle either on show at markets or at pasture. Similarly, participation at livestock markets reflects the significance of the autonomous farmer consistent with good farming. Providing measures of good farming in relation to animal disease may therefore help cattle purchasers identify good farmers, and help them avoid becoming a bad farmer as a consequence of poor cattle purchases. The extent to which such measures can successfully symbolise the good farmer and influence cattle purchasing is explored in the remainder of this paper.

Studies of behavioural influences in disease management reveal two distinct methodological approaches. On the one hand, agricultural economists, drawing on methodologies from behavioural psychology, have conducted experiments to simulate the effects of information provision and financial incentives upon bio-security behaviours. On the other hand, sociological research has sought to conceptualise and describe in-depth farmers’ responses to disease events and policy interventions. Each has their problems. Despite the promise of the experimental approach, research participants are often students responding to hypothetical situations wholly divorced from the practical skills and situational awareness that farmers use to respond to real-life context-dependent situations . By contrast, qualitative analyses of good farming and bio-security, whilst focused on real-world policies and disease incursions, are retrospective and subject to recall and social desirability biases. Rather than adopting one or the other, we seek to develop an innovative mixed-methods approach that allows us to quantitatively and qualitatively assess the value of symbolising good farming to influence farmers’ cattle purchasing decisions to prevent bTB. The following sections firstly provide information on the importance of bTB and the relevance of cattle purchasing before providing a detailed account of our methodological approach. In the United Kingdom, bTB is the UK’s most challenging endemic disease, resulting in the premature death of approximately 35,000 cattle and costing the taxpayer in excess of £100 m every year . Managed by the government, the disease has a complex epidemiology involving transmission by legally protected wildlife, the culling of which for disease control purposes has raised political, social and economic challenges . 2007. Cattle movements have become recognised as an important part of the epidemiology of bTB. Studies have shown how the movement of cattle is one of the most important risk factors in infected herds, whilst movements also translocate disease from areas of high to low prevalence . Whilst infected farms are restricted by law from buying or moving cattle on or off farms, all other farms are free to act as they please. Nevertheless, the limitations of diagnostic tests and their frequency mean that these movements still pose a risk to other farmers.

Indeed, other countries with successful bTB eradication schemes, have governed the movement of all cattle between areas of different epidemiological risk using statutory and/or voluntary policies of ‘risk based trading’ and in doing so identify and provide cultural capital to good farmers. Whilst no such scheme currently exists in the UK for bTB, policy makers view cattle purchasing as an important practice on which to apply the behavioural sciences in order to govern cattle movements through behavioural nudges rather than regulation. To understand the impact of different ways of objectifying good farming,sawtooth greenhouse we devised a novel mixed-methods approach. Avoiding experimental approaches involving non-farmers, our approach involved simulating cattle purchasing with farmers who buy and sell cattle. Many studies within the behavioural sciences involve randomised controlled trials, but this approach was not available and not suitable: we were not able to alter the information provided at the point of sale.The diversity of cattle, buyers and sellers also makes controlling for the effect of a single intervention a significant methodological challenge. Instead, our approach sought to simulate cattle purchasing, whilst also allowing farmers to reflectively deliberate on the reasons for their purchases and the value of different behavioural insights. To do this, we developed cattle purchasing game in Mural – a web-based interactive whiteboard – in which participants moved around a Monopoly-style board . Players progressed around the board by rolling one die. All games were played online via Zoom due to Covid 19 lockdown restrictions. Game play was organised using a “branch and bottleneck” structure. Branches reflect different contextual influences that participants land on at random throughout the game. This allowed us to introduce an element of competitiveness between players: points were awarded for landing on squares that reflected ‘positive’ contexts. No points were awarded for landing on negative blue squares. Red squares were a bTB test: if players landed on these, they were required to roll an even number to pass the bTB test,otherwise they would miss a go. Bottlenecks were cattle purchasing events that all players had to complete at the same time and were located in each corner square of the game board. Once one player reached a corner square, all other players also moved there.

Players were then read a cattle purchasing scenario and asked to make a choice between four adverts.They provide opportunities to elicit attitudes and beliefs about complex and potentially sensitive situations and to examine how people may respond to future events . Scenarios work best when they are based on plausible and familiar situations . Scenarios were therefore developed based on a prior research project on cattle purchasing involving farmers and vets. To ensure the scenarios reflected real-world cattle purchasing opportunities, specific versions were developed for three sectors: dairy, store cattle, and calf-rearing. For each scenario, adverts contained information to symbolise good farming in order to influence purchase choices. Firstly, farmers could use pictures of the animals to derive good farming information. For scenarios 1 and 2, pictures were of cattle in a livestock market, but for scenarios 3 and 4 animals were pictured on farm. Secondly, adverts featured two different conceptual measures of good farming. All adverts contained a logo stating how many years free the herd had been from bTB and the geographical average years free for the area in which the farm was located. Values were set randomly. In this method, good farming is symbolised by longer periods of disease freedom; ‘bad farmers’ would avoid purchasing from farms who had recently had an outbreak for fear of introducing disease. Scenarios 2 and 4 also contained a ‘Good Farmer Rating’ to graphically indicate the percentage of satisfied previous customers for each vendor. The aim of this logo was to convey levels of trust and reputation of the seller, which had seen to be important considerations when purchasing cattle from our previous research, and found in other research by Hidano et al. . Presented as a star rating, the logo was similar to review ratings found on internet shopping sites. Two ratings were set at 95% and two at 70% satisfaction. In addition to these measures of good farming, scenarios 1 and 3 explored the effect of different compensation regimes upon purchase decisions. Two different schemes were presented: two sales adverts stated that the purchaser would receive 50% compensation if the animal ever tested positive for bTB in future.

Especially system functions in arable systems were perceived to be moderately to strongly affected

Enabling conditions in the social domain were e.g. related to rural demographics and/or availability of labor and more horizontal and vertical cooperation and social self-organization . Specifically, in BGArable and RO-Mixed emphasis was put on enabling conditions in the institutional and social domain. In all case studies, interacting thresholds across level and/or domain were observed . More details on the interacting thresholds are presented in the Supplementary Materials 3. Common interactions between critical thresholds occur between field environmental and field-economic, from field-economic to farm economic, from farm-economic to farm-social, from farm-social to farming system-social, and from farming system-social to farm-social . Generally, an environmental issue at field level, for instance, decreasing soil quality , pest diseases , wildlife attacks , or drought is so much of a shock or stress that it leads to yields that are too low to sustain an adequate level of farm income . In a majority of the farming systems, high input prices and decreasing output prices and sales further diminish the farm income. Too low incomes at farm level were in all case studies resulting in reduced attractiveness of farming, farmers quitting or the lack of finding a successor for the farm. In UK-Arable, also reduced farmer happiness due to lack of recognition was mentioned as a reason for quitting a farm. Farmers quitting their farm without having a successor was in multiple farming systems also considered to contribute to a smaller rural population at farming system level . Interestingly,hydroponic nft although socially oriented function indicators and resilience attributes were less often formally included in the discussions, they eventually appeared when explaining how challenges impact the farming system. Having less farms in the farming system was also associated with a lower maintenance of natural resources and a less attractive countryside.

Interactions with critical thresholds in the environmental domain at farm and farming system level were mentioned in a few other case studies. In NL-Arable, at farm level in the environmental domain a narrow rotation in which starch potato is grown every second year was expected to lead to increased pressure of plant parasitic nematodes . In UK-Arable, low income at farm level was expected to lead to declining soil health at field level . In IT-Hazelnut and SE-Poultry, environmental regulations were expected to improve the maintenance of natural resources at farming system level, but also to push farm income levels below a threshold through increased costs . Overall we observed that environmental thresholds certainly feature, but differ in the level at which they play a role and in what direction they evolve. In farming systems for which access to land is an issue , quitting of farmers may also be an opportunity, provided land becomes available on the market for sale or to be leased. In ES-Sheep, quitting of farmers was experienced as a serious issue. In IT-Hazelnut, the retention of young people on the farms was specifically mentioned as something that could support the rural life and vice versa . Both low economic viability at farm level and low attractiveness of farming and a smaller rural population were considered to reduce the access to labor at farm level in BG-Arable, SE-Poultry, PLHorticulture, DE-Arable&Mixed, RO-Mixed, and ES-Sheep. Access to labor in BG-Arable, PL-Horticulture and RO-Mixed was important for the continuation of activities on farms, as lack of labor was expected to push yields below acceptable levels . In BG-Arable lack of labor could be overcome by implementing new technologies, but this would require a labor force with higher levels of education and qualification which is even harder to find. Lack of labor was also expected to push production costs beyond critical thresholds in SE-Poultry and RO-Mixed. Hence, in multiple systems, low economic viability, attractiveness of farming, rural depopulation and low level of services at farming system level, and low access to labor seem to be part of a vicious cycle. Following from Fig. 1, it can be made plausible that after exceeding critical thresholds of challenges, a decline in performance of system’s main function indicators and resilience attributes was expected by workshop participants in most case studies . Across farming systems, the functions “Food production”, “Economic viability”, and the “Natural resources” were in most cases expected to decline moderately or strongly .In ES-Sheep, ongoing decline of function performance was expected to be aggravated.

When discussed in case studies, “Biodiversity & habitat” and “Animal health & welfare” were on average expected to be less impacted compared to other functions. When exceeding critical thresholds of challenges, also a decline in resilience attributes was expected in most case studies, mainly because of a decline in profitability, production being less coupled with local and natural capital, a declining support of rural life and lower levels of self organization . By contrast, participants in BG-Arable and SE-Poultry generally expected improvements in resilience attributes after critical thresholds are exceeded . For instance, infrastructure for innovation was expected to develop positively in BG-Arable and SE-Poultry, while it was expected to develop negatively in other case studies . In the case of BG-Arable, participants expected increased collaboration, leading to innovation, in case the system would collapse. In the case of ES-Sheep, participants expected that the current low profitability of farmers will not allow investment in new infrastructures for innovation. All studied farming systems were perceived to be “close” or “at or beyond” at least one critical threshold for challenges, function indicators or resilience attributes . The actual state of the system may be more or less close to a threshold than the participant’s perception. Obviously, for case studies that are perceived to be “at or beyond” critical thresholds while still continuing business as usual, the actual state must be at a different position than perceived. Still, perceived closeness can be seen as a clear stress signal, indicating that change is needed, expected or even already experienced. An example refers to the ban of crop protection products before alternatives are available. This stress signal could instigate a study about a reasonable time to phase in/ out regulations regarding the use of crop protection products before actually implementing them. Perceptions of being close to or at critical thresholds also indicate that, from the perspective of farming system actors, immediate action is needed to preserve the farming system or guide it in its transition, thus avoiding a situation where sustainability is even lower. Looking at multiple challenges puts individual challenges into perspective.

To give an example, climate change may be a problem causing regime shifts in many socio-ecological systems , but for the studied farming systems this is not the only challenge and often also not perceived to be the most urgent, except for some arable systems . This supports the notion that climate change should be studied in the context of other drivers . At a global level, reducing anthropogenically induced climate change is, of course, urgent and agricultural systems’ contribution to it must be reduced. Some challenges experienced by FS actors, especially farmers, may also be implicitly caused by climate change; for instance changing legislation and high input costs. For most of the farming systems in our study, climate awareness of some stakeholders, such as conventional farmers, is however not likely triggered due to the impact of climate change on their system per se. When deliberated in an appropriate manner with those stakeholders, new legislation in the context of fighting climate change may however have considerably more effect regarding changing stakeholder perceptions. Function indicators for food production and economic viability were often perceived to be close to critical thresholds. This confirms the need to closely monitor economic indicators as is done in the CMEF of the CAP . When discussed, social function indicators were generally perceived to be “not close” or “somewhat close” to a critical threshold, except for ES-sheep where participants experienced that a critical threshold was exceeded . Environmental function indicators were in most cases perceived to be “not close” or “somewhat close” to critical thresholds . Only in arable systems,hydroponic channel environmental functions were experienced “close” or “at or beyond” critical thresholds. This was mainly related to the capacity of soils to deal with an excess or lack of water, often due to climate change. Participants in workshops of arable systems indicated that a lot of effort was already required to maintain rather than to improve the current soil quality. Arable systems, in need for soil improvement to avoid critical thresholds, would benefit from enabling conditions at national and EU level that foster the maintenance of natural resources.

Mitter et al. , based on a mechanistic scenario development approach for EU agriculture, expect improved attention for natural resources only in a scenario following a “sustainability pathway” out of five possible future scenarios. Current conditions and their future development hence do not seem to support a resilient future of arable systems. Overall, perceived closeness to critical economic thresholds could explain the perceived lower importance of social and environmental functions compared to economic and production functions . Defining critical thresholds seemed most difficult for resilience attributes . According to Walker and Salt it is actually impossible to determine critical thresholds for resilience attributes because they all interact. However, function indicators also interact, but were easier to assess for participants. We argue that difficulties in determining critical thresholds are probably more an indication of the perceived redundancy of resilience attributes for system functioning: presence and contribution to resilience was low to moderate according to stakeholders’ perceptions . This could be related to a control rationale , in which keeping a relatively stable environment and improving efficiency is more important than increasing the presence of resilience attributes. It should be noted, however, that participants often could indicate enabling conditions that improve the resilience attributes. This could be an indication that participants are aware of the importance of resilience attributes, but are in need for more concrete, locally adapted indicators that represent the resilience attributes. In any case, suggesting improvements for resilience attributes could be seen as an implicit acknowledgment by participants that building capacities for adaptation or transformation is required. Perceived thresholds may be different than the real threshold. For the systems that are perceived to be “at or beyond” critical thresholds, it is not necessarily too late to adapt in case the real threshold is actually at a different level than the perceived one. The extensive sheep system in Spain was judged to be close to a collapse, but alternative systems and strategies to reach those have been proposed . In IT-Hazelnut, introduction of new machinery in the past has made farming more attractive for the younger generation, thus avoiding depopulation . Further developments in IT-Hazelnut regarding local value chain activities at farming system level rather than farm scale enlargement, are aimed to further stimulate economic viability and the retention of young people in the area . In PL-Horticulture, the case study is relatively close to Poland’s capital where access to land is limited, system actors aim at increasing the economic viability via vertical and horizontal cooperation at farming system level, which keeps re-attracting seasonal laborers from nearby Ukraine, where wages are lower, to the region. The common factor in these examples of adaptation is that resources are needed to implement them. Be it financial, human, social or other forms of resources. The examples above also suggest that coming back to a desired state, even after exceeding a critical threshold, is possible, provided the disturbance causing the exceedance does not last too long , and adaptation strategies are available . The notion of a critical threshold being a combination of magnitude and duration was not discussed much in the workshops but could help to further define critical thresholds. For instance with regard to the number of years the farming system can deal with extreme weather events as was done in NL-Arable. It is worth noting that challenges are perceived to be more often “at or beyond” perceived critical thresholds than function indicators and resilience attributes.

This is not to suggest there is nothing to gain from using some smart farming technologies

Looking more closely at the emergence of smart farming developments, another crucial consideration pertains to the innovation processes that try to generate and integrate new and striking configurations of firms, farmers, and research institutes. New networks of diverse stakeholders are formed; extant networks are re-made, potentially rendering invisible the role of some actors and their interests. In a sense, then, smart farming can resemble the notion of “innovation by withdrawal”,which departs from the view that innovation is “structured around the introduction of a new element, an artefact, a way of operating, a service, and its success is dependent on the number of adopters and the significance of the entities which are articulated with it” . Yet, as demonstrated by the case of no-till farming in France , withdrawal of one element relies on making visible hitherto invisible or overlooked elements , while maintaining problematic practices.To respond, an alternative to focusing on introduction or withdrawal is to recognize that agricultural innovation, like any other practice, is always a topological affair: it is about overseeing and managing configurations of humans and materials and how they flow through a system or across a specific domain, such as a field . Where there are blockages, conduits can be installed to increase flows; where there are leaks, plugs are required. Farmers shift and prod to adjust arrangements of materials or relations with a view to addressing problems, such as falling yields, vulnerabilities to climatic variability, or exposure to viruses. A mechanism of implementing innovations such as those associated with smart farming is to reconfigure topological arrangements. As demonstrated by literature on agricultural biosecurity,farming practice needs to be viewed as occurring against the topological backdrop of an “entangled interplay” , with numerous “contingent intra-actions” occurring across multiple risky “borderlands” .Proximities,distance,nft system connectivities,and modulations of presence/absence figure in the effort to create desired outcomes,with new insertions or removals pursued in efforts to control or steer activities in defined or experimental ways.

Whether the risk is a matter of falling yields or exposure to viruses, it pays to acknowledge the ongoing relationship between farming and innovation through a topological lens; that is, to dwell on farmers as active agents of topological transformation, even if they are rarely acting alone. One way to combine an analytical concern with digital life, corporate interests in establishing a certain type of smart farming, and topological transformations is highlighted by literature on smart farming innovation processes. Consider, for example, how new configurations come into the picture. As highlighted in research on smart farming in Canada, and as referenced by Relf-Eckstein et al.,the Canadian government has recently established innovation ‘superclusters’ to examine and exploit technological opportunities. A recent outcome is an industry-led consortia called Protein Industries Canada, which includes a partnership whereby Lucent BioSciences “will use the hulls of pea and lentil seeds which are a co-product from value-added processing completed by AGT Foods and Ingredients [to create] Soileos: a novel carbon-neutral micro-nutrient fertilizer that uses organic fibre as a carrier to provide micro-nutrients to plants” . As this case suggests, the ‘smart’ in smart farming can involve astute and imaginative arrangements to make new products and chase after profits in novel ways. A similar picture emerges in the Netherlands where a “golden triangle” of agricultural research, industry, and government aims to create “new business ecosystems consisting of focal firms, their suppliers, complementor firms, and customers” . A key feature is the leading role of the Dutch firm Philips, which occupies a prominent position in high-tech urban agriculture , a growing smart farming sector, by “providing the essential technologies, registration of patents, and creation of new business opportunities” . Meanwhile, in the larger and more traditional Dutch agricultural sector, the Food Valley Open Innovation Ecosystem includes “the Wageningen Campus and the planned World Food Centre in Ede” and creates ties between research and development centres run by large firms such as Friesland Campina and Unilever and wider networks of small-to-medium enterprises and startups.

There are 15,000 scientists across Food Valley, with twenty research institutes, 1440 food related, and 70 science related firms . Such configurations of firms, farmers and research institutes will likely create new smart farming products and services and build on Dutch successes in exporting around €9 billion worth of high-tech agrifood innovations, including “energy-efficient greenhouses, precision agricultural systems and new discoveries that make crops more resistant” . Put differently, the topologies of smart farming point toward new forms of “path creation” that involve but also often extend beyond farmers. It is instructive that smart farming today is bound up with efforts to use ‘open innovation’ processes that facilitate co-design or co-innovation between agricultural technology providers, farmers, and others. Such an approach can “blur the boundaries between scientists and agricultural system stakeholders, between agronomists and farmers, and between actors in the agricultural sector and those designing in other sectors” . The virtue of “participatory design processes involving farmers” is that it can yield new tools, such as dashboards , to help farmers understand agro-ecological conditions. There are, however, no guarantees that smart farming developments will yield effective configurations. Indeed, there is significant evidence that smart farming developments are hamstrung not only by the instrumental logics underpinning technology providers but also by ineffective coordination and inadequate arrangements of materials or skills. Consider here the push to develop automated body condition scoring and a soil water outlook tool for Australian dairy farmers. Noting that these versions of smart farming innovation involve “a unique innovation challenge [not least because of] the new knowledge demands for farmers in a highly dynamic, technology-driven environment” , one finding is that the new tools and practices confront limitations in the way agricultural relations are configured with respect to the wider institutional milieu. Making the most of the soil water outlook tool, for example, required but did not receive sufficient input from the Australian Bureau of Meteorology “to help farmers to link the SWO with seasonal climate outlooks” . Then, with regards to automated body condition scoring, the new technology led some farmers to think “maybe we don’t need the [farm] advisor as often,”,with the upshot that “some tools were potentially replacing the skills of advisors”.Yet, because “more remote monitoring of key performance indicator data via online software” can enable farm advisors to make fewer farm visits, smart farming in this context conceivably increases the sense of isolation many farmers already experience .

Elsewhere in Australia, smart farming developments call attention to a different dynamic between farmers and advisors. In some rice farming regions, advisors might be expected to be the “sense makers” who can explain and encourage farmers to adopt new technologies; but in fact one consultant respondent claimed “it’s mainly been the farmers dragging the agronomists along” .At the same time, “insufficient support structures” , for example regarding data compatibility or standards, can hold back adoption and frustrate farmers who are “prepared to use evolving and uncertainty-generating technologies” but find that their knowledge is not effectively tapped. In Canada, ‘broadacre’ smart farming developments occur amid the “critical constraint” of labour shortages and demographic change, but even here “adoption is lower than anticipated” , with one explanation focusing on tensions around what happens to data produced on farms. A problem yet to be overcome is industry self-regulation of data usage and a lack of certainty about the legal ramifications of smart farming. Thus, “[u]ntil clarity is brought to the issue of data, the industry is at risk of losing farmer’s trust and potentially hindering innovation opportunities at the farm level” . Although there are examples from the literature which demonstrate that smart farming innovation involves an ongoing process of trying to reconfigure arrangements of sociotechnical relations, I argue a more accurate and urgent conclusion is to emphasize the ‘mis-configured innovations’ of smart farming. One of the main features of smart farming concerns the limited parameters within which innovations operate. In Canada, for example, an element in smart farming arrangements is models and platforms designed for commodity farmers, not those “farmers working outside of the dominant industrial model” . In effect, “the maps created within those big data platforms developed by industry are made meaningful only if one adheres to a rigid conventional farming strategy of seeding in neat rows separated by areas of soil free of weeds” . A similar result emerges in Australia where observers note that farmers want autosteer technologies, new imagery services, levelling and GPS guidance because “if they’ve got efficient layouts, laser levelled, they’ll make significant water savings and they’ll have reduced labour inputs as well” . Smart farming therefore means that food producers contemplate, “standardizing the environment” in accordance with the commercial imperatives of farmers operating large holdings and using expensive machinery to generate predictable topographies that fit with the new topologies required to make smart farming technologies effective. Built-in biases pervade all algorithmic systems ; the biases in smart farming might only pertain to environments in the first place but they can have broader political-economic effects.

As such, the core problem with the various reconfigurations underpinning smart farming developments is not simply that the absence of one or other action or reform can limit their impact,hydroponic gutter but rather that smart farming innovation processes begin and proceed without adequately conceptualizing the underlying obstacles and limitations confronting food producers today. Technological innovations that reinforce power asymmetries regarding data ownership, for example, or that fail to challenge implicit biases toward certain types of environments, render some interests invisible while reifying specific types of logics, such as narrow measures of economic efficiency. Like any innovation, insertion, or reconfiguration, smart farming entails topological transformation; but problems emerge when the “quieter registers” of smart farming make it possible for “powerful actors to make their presence felt at one remove, to reach into the everyday life of distant others” , for example by dispossessing them of valuable data or establishing algorithmic biases toward standardized farm topographies.It can make sense to use devices or services in new arrangements that create efficiencies or give food producers new access to information that can inform decisions. However, because these developments always by necessity involve reconfiguring arrangements of sociotechnical relations, agricultural innovation processes will continue to introduce new misconfigurations when they pursue discrete solutions to specific problems, rather than integrated developments based on incremental adjustments in information-intensive iterative processes that target systemic or structural change. As insisted upon by scholarship on food sovereignty in critical agrarian studies , the urgent challenge today is to conceptualize a planetary land, agrarian, and food system in which food producers and consumers everywhere are confronted by, but examine ways of overcoming, the same problems of neoliberal capitalism dominated by transnational corporations, authoritarian governance, and climate change. In the shadow of the corporate food regime, producing food in the Netherlands or Canada is bound up with the realities of producing food in India or Kenya. Further, the dynamics of digital life mean smart farming innovations in one place will inform and conceivably move the ‘planetary cognitive ecology’ generally, with unpredictable but connected results playing out elsewhere. The products of smart farming will only reinforce problems if they yield new patents for agricultural technology providers in a place such as Ireland , a widening yield gap between capital- and labour-intensive agrarian systems , or if they increase the likelihood of ‘smart’ food production in one region leading to food dumping in another . Per the vision of a Common Food Policy in the European Union , rather than seeing smart farming developments “reinforcing existing production models, leading to trade-offs between different environmental impacts, or between environmental and social sustainability,” the task is to reorient innovation “towards low-input, diversified agroecological systems” . In the light of these challenges, a sustainable and successful smart farming innovation process requires what we might imagine as the coproduction of ambitious ‘topological repertoires’ that make ongoing assessments of absence, presence, proximity, and reach at the scale of a structure or system and then pursue appropriate technological solutions from the ground up.

All three contracts result in farmers increasing rice area relative to control farmers

Examining results of farming contracts on the other three variables of interest, we also find consistently positive and significant effects. Focusing on the ANCOVA estimates with covariates, being offered a farming contract increases yields by 473 kg per hectare, a 29 percent increase in yields compared to the control. Given that we offered three types of contracts, this result does not immediately reveal what contract attributes most contributed to the yield gains. What is clear is that farmers did not simply fulfill their contracts by increasing the amount of land planted to rice. Rather, their productivity per unit of land increased in response to signing a farming contract. Not unexpectedly, farmers with production contracts increase their market participation by selling 35 percentage points more of their rice harvest, a 140 percent increase above farmers without contracts. This result may appear tautological, as farmers with contracts are expected to sell the contracted quantity to ESOP. However, even with contracts, farmers sell well less than 100 percent of their rice crop, implying that farmers produce enough rice to meet the terms of their contract and are able to decide how to dispose of the excess quantity, either by consuming the rice or saving it as seed for next year.One concern in the existing literature on contract farming is that by signing a contract, farmers reallocate land and labor to the contracted crop. Thus, while farmers may increase their production on one crop, the overall income effect may be zero or negative . We find that farmers in the treatment earned $140 more per person, an increase of 52 percent or about four tenths of a standard deviation above the mean for control households. This is a substantial income gain in a country where GDP per capita is around $800. In considering how farmers increased yields and income, Table B3 in the Appendix presents results from ANCOVA estimates of treatment on seed, fertilizer, pesticide, herbicide, and labor.

Treatment significantly increases the use of each input, indicating that the contracts resulted in an intensification of rice cultivation,mobile grow rack in addition to the extensification show in the regressions of rice area on treatment.19 Overall, our results, the first from an RCT, provide consistent evidence that contract farming has a positive and significant impact on several measures of farm productivity and household welfare. At least for rice growing households in Benin, contract farming appears to be a mechanism that encourages vertical coordination and can contribute to rural transformation.Given these positive results, it is particularly important to understand which contract attributes matter most in increasing yield and income. To do this, we randomly assigned treated households into one of three contract types. Fig. 3 summarizes the effect of each of the three types of contracts by drawing distributions of post-experiment values for each outcome. To the distributions we add vertical lines to mark the unconditional mean for each outcome variable by contract type. Visual inspection shows some heterogeneity in outcomes based on contract attributes. We also present regression analysis of these treatment effects, which not only allows us to test for differences between each treatment and the control but also test for differences between one treatment and another. Results from these regressions are presented in Table 6, with Bonferroni-adjusted Wald tests for differences between co-efficients on the treatment dummies in Table 7.However, testing for differences between the magnitudes of the co-efficients reveals that the effect of T1 is not significantly different from the effect of T2 or T3 . By comparison, the effect of T2 on area planted to rice is significantly lower than the effect of T3. While one could expect that the provision of input loans lowers the per unit cost of production, allowing farmers to expand area planted to rice without increasing their total farm production costs, it is less obvious why farmers with a contract that only guaranteed a price planted a similar sized area.

It may be that farmers who were to receive the extension services decided to focus effort on applying their training to a more circumscribed area. For those in T3, the addition of the input loan to the training may have reduced costs enough for farmers in this group to increase their area planted to an amount similar to those in T1. However, we lack the detailed farm production data needed to test this hypothesis. Turning to each contract’s effect on yield, we find that all three have a positive and significant impact. The magnitude of the impact varies slightly, from about 450 kg per hectare for farmers in T2 to about 500 kg per hectare for farmers in T1 and T3. A Wald test for differences between each of these co-efficients fails to reject the null of equality . One possible explanation for this results is that, given the variance in yields, we lack power to detect significant differences across treatment arms. A second possible explanation is that farmers gained little in terms of productivity by receiving extension training or input loans. Table B4 in the Appendix provides ANCOVA estimates of each contract on input use. Each contract significantly increases seed, fertilizer, and labor use, though contracts tend to have a null effect on pesticide and herbicide use. Bonferroni-adjusted Wald tests for differences between co-efficients on each contract indicator are never significant, indicating that across treatments farmers used about the same level of inputs. While far from conclusive, we take this as suggestive evidence that simply resolving price risk was sufficient to allow farmers to increase their use of inputs and thereby substantially increase yield. All three contracts have a positive and significant impact on market participation. However, unlike yield, in which each contract’s effect size was statistically similar, the impact of each contract on market participation significantly differs from each other. Conforming with our priors, effect sizes are greater for contracts that offer more services to the farmer. Those in the T1 treatment sell just under 50 percent of their rice harvest into the market , while those in T2 sell 57 percent and those in T3 sell 66 percent. The effects of using contracts to integrate farmers into the market are clear.

Without a contract to produce rice, households sell about a quarter of their rice production and keep the remaining three quarters. Under the most complex contract, farmers nearly reverse this ratio, selling almost 70 percent of their rice into the market and retaining only 30 percent. The evidence for each type of contract’s impact on income per capita is less obvious than when we simply compare all contracts to no contract, as in Table 5. While the effects of all three contracts are positive relative to the control, only the effect of the input-supply contract is consistently significant. Yet, when we conduct Wald tests for differences between co-efficients, we consistently fail to reject the null of equality. We speculate that this is due to a lack of power sufficient to detect differences in effect size in a notoriously noisy variable such as income and not evidence of a true null. We base this on the similarity in the size of co-efficients and standard errors across the income regressions in Tables 5 and 6. Overall, we find a curious degree of variation in impacts based on the terms of the contract. Contrary to our priors, it is not always the case that the effect size of T1 is smaller than T2, which is smaller than T3. Instead, we find that the fixed-price contract increases rice area to the same extent as the input-supply contract , while the production management contract has a smaller effect. All three contracts have similar effects on yields, meaning that the provisioning of extension training and/or input loans does not result in increased yield relative to the contract the only provides a price guarantee. For income per capita we again find that the added elements of T2 and T3 do not seem to provide much additional value over the simple fixed-price contract. Throughout the analysis, we frequently find that the magnitude of the co-efficient on the T2 treatment is the smallest of the three treatment arms, while the magnitude of the co-efficient on the T1 treatment is only slightly less than that on the T3 treatment. In fact, the only outcome variable that conforms to our prior is market participation, where farmers with the production-management contract sell significantly more rice than farmers with the fixed-price contract, and farmers with the input supply contract sell significantly more rice than the other two.Columns display ANCOVA results for the four outcome variables as the dependent variable. Each row designates which covariate is interacted with the treatment indicator. Cells report the co-efficient and standard error on the interaction term of household covariate and treatment indicator on the dependent variable .

We find almost no evidence of heterogeneous treatment effects by baseline characteristics. We fail to reject the null that any of the covariates mitigate or accentuate the effect of contract farming on the area of land put into rice production. For yields and market participation, a marginally significant degree of heterogeneity exists based on a farmer’s previous training in rice production. For income per capita, the only interactions that are significant are household size with the contract and experience producing rice with the contract. In both cases, larger households and more experienced rice producers had lower income with the contract than similar households without the contract. To provide a more detailed exploration regarding these three potential sources of heterogeneity, we graph the marginal effects of each interaction term on our outcome variables. Fig. 4 plots the marginal effects and 95 percent confidence intervals for the interactions between household size and indicators for each type of production contract. Panels document the effect on one of the four outcome variables. As was evident from Table 8, there is a lack of heterogeneity in household size on rice area,ebb and flow table yield, and market participation. For income per capita, we find that smaller households offered the input-supply contract have higher income than control households of similarly small size. As household size increases, income per capita for all groups decreases until there are no significant differences across treatments. It appears that when households are relatively small , they are better able to take advantage of being offered the input-supply contract and convert it into more income for each member. This difference diminishes for households with more than eight members and is not significant for the other treatment arms. Fig. 5 presents a similar set of margin plots for the effect of experience in rice production, measured in years. As with household size, there is little evidence of heterogeneous treatment effects on rice area, yield, or market participation. Where significant evidence does exist is for income per capita. Without a contract, more experienced farmers have higher income than less experienced farmers. This heterogeneity based on experience disappears for farmers randomly assigned to a production contract. Regardless of the type of contract, less experienced farmers have approximately the same amount of income as more experienced farmers. Contract farming helps inexperienced farmers earn incomes comparable to that earned by much more experienced farmers. It takes farmers without a contract a decade or more of experience to earn similar levels of income. Finally, Fig. 6 graphs the marginal effects of each treatment interacted with an indicator for whether or not the farmer had participated in training in the last 12 months. Because the household characteristic is now a binary variable, we graph each contract along the horizontal axis and the lines represent if the farmer participated in training. Here again we find little evidence of heterogeneity. As was evident in Table 8, households with training and a contract had higher yields and greater market participation than households with training in the control. But there are no significant differences across treatment arms and no significant differences within treatment arms across training/no-training. To some extent, our heterogeneity analysis appears fruitless. Across a number of different pre-specified covariates and a number of different outcome variables, we fail to find much evidence of heterogeneity in treatment effects.

Plant diseases can potentially prevent grain harvesting entirely in severe circumstances

The initial step for Soil preparation is testing the soil. It involves identifying the soil’s current nutrient levels and the suitable amount of nutrients to be feed to a certain soil based on its fertility and crop demands. The values from the soil test report are being used to categorize a number of key soil parameters, notably Phosphorus, Potassium, Nitrogen, Organic Carbon, Boron, as and soil ph. Irrigation is a type of agriculture that plays an important role in water and soil conservation. Complicated data could be used to maintain irrigation performance and consistency when assessing systems with respect to water, soil, climate, and crop facts. Weeds are plants that is grown where it is not needed. It includes plants that are not intentionally sown. Weeds compete for water, nutrients, light, and space with agricultural plants, lowering crop yields. Weeds can diminish the commercial worth of agricultural regions by lowering the quality of farm products, causing irrigation water loss, and making harvesting machinery harder to run. To control weeds, farmers often spray homogeneous herbicide spraying throughout the field twice or three times during the growth season. However, this method has resulted in the uncontrolled use of large volumes of herbicides, which is harmful to humans, non-target animals, and the environment . Plant diseases can have a devastating influence on food safety, as well as a considerable loss in both the quality and quantity of agricultural goods.As a result, in the field of agricultural information, computerized identification and diagnosis of plant diseases is widely needed. Many approaches for doing this problem have been offered, with deep learning emerging as the preferred method because to its excellent performance. Hence this work focuses on the steps involved in cultivation of crop.

It uses Deep Learning and Machine Learning algorithms to deliver solutions to various challenges faced during cultivation. It mainly focuses on recommending the crops based on weather parameters, stacking pots suggesting the nutrients requirements and specifying the Growing Degree Days. It also helps in identifying the weeds and recommending herbicides for the same. Many insects ruin the crops hence pesticides are recommended based on the insects that are present in the field. And finally cost estimation is very much needed in these recent times. Crisis, uncertainties would result in great loss. Hence forecasting the cost for cultivating a crop is necessary to plan for future uncertain events. This work specifies various costs in cultivation for future years. Crop growth is primarily influenced by the soil’s macro-nutrient and trace mineral content of the soil. Soil being the broad representation of several environmental factors including rainfall, humidity, sunlight, temperature and soil ph. The use of a support vector machine and decision tree algorithm to distinguish the type of crop based on micro-nutrients and meteorological characteristics has been presented as an efficient means of predicting the crop. Three crops where selected such as rice, wheat and sugarcane. Based on certain observations details about micro-nutrients where been obtained. These details where feed into the classifier model that in turn predicted the crop based on the passed values. There are many Machine Learning algorithms that works in a different manner. Hence selecting only two models will not provide the required output. The accuracy score of SVM was greater than decision tree algorithm with a sore of 92%. In this work best out of two algorithms is selected. But there are various algorithms dedicated for classification tasks. There is a need for working on other models such as K Neighbors classifier, Logistic Regression, Ensemble classifiers. These algorithms are indeed applied in proposed research work. The predicts only a crop based on the values entered into the SVM model. Data is most valuable. Hence more information can be obtained apart from using them for prediction. The proposed research work not only recommends the crops and also uses the data to obtain various information that would provide a detailed view about the predicted crops this includes specifying the Growing Degree Days such as heat units, amount of heat needed for the crop growth and the amount of nitrogen, phosphorous and potassium content need to be supplied for the growth per 200 lb. fertilizer. Machine Learning algorithms such as SVM and decision tree classifier was used but in this work Machine Learning algorithms such as Decision Tree, K Nearest Neighbor, Linear Regression model, Neural Network, Naïve Bayes and Support Vector Machine was used for recommending a crop to the user.

It has provided an exposure to other algorithms compared to. Linear Regression model was used to predict the production value against the climatic parameters such as rainfall, temperature and humidity. The scores of all these algorithms were below 90%. This work was just a model implementation using the data set. Web interface needs to be implemented so that even common people can use it efficiently. All the values need to be provided manually for the model to predict the crop. The proposed work helps in extracting temperature and humidity values using Web Scraping. Hence manually entering the values are not needed. The proposed work provides an interactive web interface where the user specifies the average rainfall and soil Ph value. The temperature and humidity details are extracted automatically and feed into the best model that includes 10 algorithms with hyper parameter tuning. The proposed work tends to achieve an accuracy of 95.45% with hyper parameter tuning the algorithms which was not included in. The predicted results along with certain information are displayed in the web interface which makes the user to understand the results more efficiently. Base temperature of a given crop can be used to calculate the GDD Growing Degree Days. The main aim of this study is to come up with easy and mathematically acceptable formulas for calculating GDD’s base temperature. Temperature data for snap beans, sweet corn, and cowpea are used to propose, prove, and test mathematical formulas. These new mathematical formulae, in comparison to earlier approaches, can produce the base temperature quickly and correctly. These formulas can be used to calculate the GDD base temperature for every crop at any developmental stage. This work provides a formula to calculate the GDD for the crops. Hence the formula specified in was applied to the predicted crop to estimate their GDD in the proposed work. Weeds grown along with soybean can be detected using K-means and CNN model. K-means were used for identifying the features of the images and convolutional neural network for was used for classifying the weeds and soybean. It also suggests that accuracy can we improved by fine tuning the CNN model. CNN model provides an efficient way to detect the weeds present among crops. When used along with Kmeans initially the images and its augmentations are clustered and on using CNN model helps to precisely identify the weed. The proposed work uses the pretrained model such as Resnet152V2 hence it has important layers such as skip layer and identity layer. The main goal of these layer is to make sure that the output image is same as the input. This increases the accuracy and the predictions are correct.

Not only predicting the image the proposed model also helps to provide details about the herbicides that can be used which is an additional information for the user. Existing deep learning techniques are used for weed detection. This study provides information of various ML and Deep Learning algorithms that can be used for identifying weeds. It mainly emphasis on pre-trained models. It suggests that pre-trained models as lot of benefits and hence can be used to image classification. It also provides guidance of how to work on datasets and make the datasets efficient for building the models. Many public datasets are available on various platforms that can be used for this purpose. It specifies Image Resizing, data augmentation, image segmentation some of the techniques would bring about accurate classifications and tendency of increasing the accuracy is also more in pre-trained models. Since this study provides directions to perform deep learning techniques the proposed model has opted certain techniques preprocessing steps such as Image Resizing, data augmentation is opted before building the actual deep learning model to predict the weeds. Another algorithm that can be used for identifying weeds in vegetable plantation is the CenterNet. CenterNet is used for weed identification. It includes two stages. In first stage the Bok choy images were collected and detected. In the second stage,grow lights color-index based segmentation were performed on the images collected to identify the weeds present in the dataset. The images were collected from Nanjing, China. The images were augmented to increase the dataset size and images were annotated. CenterNet algorithm was used for both training and testing the images. It is a ground-based weed identification technique. More optimization would lead to better results was suggested. CenterNet algorithm is simple yet there is a need an algorithm that strives to get correct prediction. The proposed work uses Resnet152V2 algorithm that strives to achieve more accuracy since it has special layers such as skip layer and identity layer that tries to get input image as output itself. Hence predictions would be absolutely correct. Hence Resnet152V2 algorithm is selected to obtain accurate prediction and based on the prediction obtain the list of herbicides. Fig. 3.1.

System architecture. Farmers face a challenging task in identifying crop insects since pest infestation destroys a substantial portion of the crop and affects its quality. The use of highly skilled taxonomists to correctly identify insects based on their physical traits is a shortcoming of traditional insect identification. Experiments were conducted using image characteristics and ml algorithms such as neural networks, support vector machine, k-nearest neighbors, naive bayes, and convolutional neural network model to identify twenty-four insects from the Wang and Xie dataset. To increase the performance of the classification models, 9-fold cross-validation was used. The CNN model had the greatest classification rates of 91.5 percent and 90 percent, respectively. The results revealed a considerable improvement in classification accuracy and computational time when compared to state-of-the-art classification algorithms. This work has used basic CNN model for classification as well as the same dataset used by various researchers. Hence the proposed model has used a different dataset called the Pests’ dataset from Kaggle website. This dataset consists of 9 classes of insects. Each image is taken from different locations. This dataset was selected for the proposed model since the model is trained of images about various locations that gives more knowledge for the model to understand the image and distinguish them. The proposed model uses Resnet152V2 model for classification. The Resnet152V2 model is the basic model and top of which Global Average Pooling 2D, Dropouts and more hidden layers are been implemented. This refers to fine tuning the base pre-trained model. This helps in extracting more information and helps in efficient classification. The association between the degree of difficulty in identifying insects and the identification key was investigated in this article. For a collection of 134 insects, the SPIPOLL database was utilized to generate 193 characteristic value pathways. Based on the average IES of all the insects with that of characteristic value was formulated. The CV’s derived IES was then used to generate an estimated IES for each bug, resulting in a ranked list of insects. Finally, the anticipated bug ranking list was compared to the actual bug ranking list. The results showed a significant correlation between the estimated and actual truth IES, indicating that the CV can be used to estimate the IES of SPIPOLL insects. This work has specified of how to consider the features of an image with respect to insects’ dataset. Its main goal is to identify a key that helps in distinguishing the classes. This proposed work contributes in specifying that a key is important for distinguishing the insect classes. Hence the proposed work uses Resnet152V2 algorithm for this very reason. Resnet152V2 is a pre-trained model and it automatically picks the important features rather than manually defining them. The Resnet152V2 base model on addition with Dropouts helps in removing unnecessary hidden layers and selecting the relevant ones is an advantage. Identification of insects does not solve the problem completely. Suggesting Pesticides provides a complete solution.

Young and educated laborers were likely to migrate to urban areas to seek well-paid work

Historically, farmers in both study areas have generally practiced subsistence farming, which is typically rain-fed farming and is strictly subject to an erosive environment. To effectively control soil erosion and improve agricultural conditions, both districts have acted to conserve the soil since the 1980s using methods such as terracing on gentle slope lands and building check dams to form quality cropland in the valleys. As the pilot sites for the GGP, both districts have participated in the slope cropland retirement program and grazing prohibition to protect vegetation from excessive disturbance since 1997. Such soil conservation activities have greatly improved dryland farming condition. Just as the results of our analysis suggested, a total of 30,900 ha of the croplands on the steep slopes in the Ansai District were converted into woodlands and grasslands under the GGP during the period from 2000 to 2017. Accordingly, the average soil erosion rate decreased from 57.6 t ha 1 yr 1 in 2000 to 27.8 t ha 1 yr 1 in 2017, while agricultural output increased from 291 million CNY in 2000 to 1434 million CNY in 2017. Similarly, the amount of cropland on the steep slopes of the Anding District decreased by 7.7%, while the area covered by woodland and grasslands increased by 19.0%, and the average soil erosion rate decreased by 47.4%. Agricultural output rose by 225.9% over the same period. The increase in agricultural output shows a similar pattern to the decrease in soil erosion; that is, less soil erosion leads to more agricultural output, indicating a synergy between soil conservation and agricultural production in both study areas. Similar findings indicate that soil conservation in areas with rain-fed agriculture significantly improves soil moisture and retains soil nutrients in arable land. At the same time, increasing the amount of vegetation greatly improves the microclimate , reduces flood levels, nft growing system and enhances the capacity to preserve water resources , leading to an increase in production.

Our results support the argument for soil conservation-oriented dryland farming, which was introduced in the Loess Plateau by several scholars at the Institute of Soil and Water Conservation, Chinese Academy of Science in 1980s . There are several such paradigms of soil conservation-oriented dryland farming in both study areas. The small catchment of Xiannangou in Ansai, for example, has been involved in several environmental programs, including the GGP, which is the largest cropland retirement project in the world, since 1997. Following the concept of a coupled socialeecological system, the ecological engineering design in the small catchment of Xiannangou consists of revegetation belts on the upslopes, terraced apple orchards on the gentle slopes, and grain and cash crops on croplands that are fed via check dams in the valleys . Evidence shows that such a pattern of ecological engineering not only enhances the capability of an ecosystem to provide services but also improves the livelihoods and environmental awareness of stakeholders . Compared to Ansai, which has a landscape of narrow and deep valleys, Anding is covered by wide and shallow valleys with an average gully density of 2.4 km/km2 . To improve local dryland farming conditions, Anding has constructed terrace on the cropland with slopes >5 since 1960s . By the end of 2017, terracing represented approximately 88% of the cultivated cropland in this area. Anding has therefore invested heavily in potatoes and medicinal plants because its relatively flat landscape and climatic conditions are suitable for such crops. The output value of potatoes and medicinal plants has consequently risen from 14.3 and 10.2 million CNY in 2000 to 24.1 and 34.8 million CNY in 2017, respectively.In addition to soil conservation, factors such as population density, labor, agricultural machinery use, and irrigation also positively contribute to crop farming. In fact, the study areas experienced a sharp reduction in the available amount of arable land following the introduction of the GGP, and crop farming has consequently become concentrated within a few flat croplands and check dam-formed croplands in which sufficient water sources are available for irrigation and soil nutrition. Farmers have therefore tried to increase financial capital input into agriculture and irrigation per unit area to offset the significant decrease in the available area of slope cropland in order to maintain their income. Although a significant increase in livestock output has occurred, there appears to be a trade off between crop farming and livestock farming. The factors that facilitating crop farming exert adverse effects on livestock farming, as indicated by the Pearson correlation in both the study areas.

The output shares of crop production to agricultural production in the Ansai District increased from 80.89% in 2000 to 90.11% in 2017, while the output shares of livestock farming decreasing from 16.11% in 2000 to 9.89% in 2017, showing a clear trade-off between livestock farming and crop farming. Similarly, in the Anding District, the output shares of crop and livestock farming accounted for 84.38% and 15.62% in 2000, and 94.54% and 5.46% of the total agricultural production in 2017, respectively. These changes were as noteworthy as those observed in Ansai, livestock farming in this area has significantly diminished. The reasons for such a trade-off between crop and livestock farming likely included the following: First, both the study areas traditionally practiced livestock grazing, in particular sheep raising, before the implementation of the GGP. However, after the prohibition of grazing due to the GGP and other environmental programs, inaccessibility to grass resources forced livestock farming to shift to house feeding that, to a great extent, has been dominated by pig or cattle breeding and has largely relied on forage crop cultivation in both study areas. Thus, to provide animal feed, livestock farming competed with crop farming for land. Second, farmers in both areas traditionally preferred to invest most of their land, money, and labor resources in crop farming. As a result, the input of land, money, and labor into animal husbandry has been insufficient. Additionally, grazing prohibition has meant that much of the grassland that was previously converted from slope cropland under the GGP was currently not utilized. Grazing prohibition contributes to the restoration of degraded grasslands, conservation of the soil, and improvement of ecological quality . However, long-term grazing prohibition reportedly produces soil biocrusts , which affect hydrological processes even biodiversity . In contrast, moderate grazing improves the soil properties of grasslands, maintaining healthy grassland ecosystems . To mitigate the trade off between crop farming and livestock farming as well as to fully utilize grasslands while implementing the GGP, animal feeding should include a combination of house feeding and grazing, together with rotational grazing through zonation. Regarding the issues of implementing rotational and moderate grazing in the Loess Hills, the determination of the grazing intensity that is capable of simultaneously maintaining the service provision of grassland ecosystems is required in future.

The Loess Plateau is underdeveloped relative to the rest of China and suffers low investment in agriculture. These issues have, to some extent, hampered agricultural development in this area. According to the Annual Yearbook of Chinese Agriculture, the average investment level per unit of cropland was 18,269 CNY/ha nationwide; in contrast, it was only approximately 7637 CNY/ha for the Loess Plateau in 2017. Additionally, the Loess Plateau is the main area from which rurale urban migration occurs in China. A report on Chinese migrant workers in 2017 by the State Statistics Bureau indicated that the number of rural laborers migrating to cities rose to 287 million, representing over 49.7% of China’s rural population, and more than half of those migrants were between the ages of 18 and 35. The Loess Plateau has suffered particularly severe loss of young labor over the past two decades and the combination of insufficient investment and rural-urban labor migration places heavy restrictions on farming and rural development , as the relations between agriculture and the variables of labor force and population density indicated in this study. Therefore, more attention should be paid to improving agricultural infrastructure and technical investment in this region. Resolving the loss of rural labor in the underdeveloped areas of China is therefore a particularly important challenge for policymakers to address. In addition, the education and skills of rural laborers are key factors for agricultural development . Given that labor is an important determinant for agricultural development, as indicated by the regression results in this study, more resources should be dedicated to rural education and skill training for rural laborers.Grasslands cover approximately 26% of the total global land area and are the most widely distributed terrestrial ecosystems on Earth . In China, grassland areas cover over 2.6  108 ha, accounting for about 27% of the country’s land area. Therefore, conservation and rational use of such grassland resources in China have been the focus of scientists in various disciplines. Grazing is the most important utilization practice of grasslands, and grazing livestock is considered one of the critical biological components of grassland ecosystems. From a historical perspective, the primary purpose of grazing has been producing meat and milk products for humans, which occurred at the expense of many other potential functions of grasslands . In recent decades this perspective has changed due to several studies, nft hydroponic system which found that grazing by large herbivores maintains the stability of grassland ecosystems and increases the multi-functionality of grasslands . Therefore, livestock grazing was determined to be an essential pathway to producing livestock products and maintaining and promoting grassland ecological service functions. Consequently, pasture-based ruminant farming systems with full- or part-time grazing have increasingly emerged as a strong option for achieving “wine win” outcomes between grassland ecological functions and livestock production functions , which have significance in promoting the sustainable development of livestock farming.

The following passages discuss grassland-based ruminant farming systems’ ability to meet current human needs for highquality livestock products and grassland ecosystem services. We provide insights into the main challenges and future scientific research directions associated with the development of grassland based ruminant farming systems.Grazing is the most economically conservative way to raise livestock because the cost of the herbage consumed by grazing ruminants is lower than that of mixed rations . Additionally, in a grazing system, ruminants feed autonomously, with minimal mechanized equipment and labor. Hence, grassland-based livestock farming systems can be highly profitable, as shown by Mwebe et al. , who found that farmers who allow grazing in small herds made higher profits than those who tried other feeding strategies. Apart from the low-cost production of livestock products, grassland grazing allows farmers to produce high-quality, niche foods with a higher market value than similar products derived from intensive, indoor livestock management. For example, higher antioxidant activity was observed in goat cheese from grazing goats due to greater quantities of polyphenols, hydroxycinnamic acids, and flavonoids consumed during grazingd resulting in goat cheese with higher bioactivity. Previous studies have also demonstrated that grazing ruminants showed improved milk quality, with enhanced sensory properties , increased mineral content , and greater odd- and branched-chain fatty acid concentrations . Furthermore, grazing also improved the meat quality of ruminants by enhancing the fat color and increasing the n-3 polyunsaturated fatty acid and vitamin E content in the muscle.Higher-quality products may result from the fact that livestock welfare is improved within grassland-based systems.Therefore, grazing systems can reduce feeding costs and produce high-quality livestock products.Livestock grazing plays a vital role in regulating grassland ecosystems functions , allowing suitable grazing systems have the potential to provide valuable ecosystem services for humans. For example, Sollenberger et al. reported that the excreta of grazing livestock was a source of nutrients for grasslands. Their study determined that managing livestock grazing to increase the uniformity of excreta deposition increased the efficiency of nutrient cycling and changed the composition of soil microorganisms and above ground plants. Several studies have shown that moderate grazing improved grassland ecological functions: boosting plant productivity and biodiversity, improving soil structure, fertility, and microbial biomass, enhancing carbon and nitrogen storage, and limiting erosion. Gong et al. found that moderate grazing resulted in higher species diversity and below ground root biomass, driving greater productivity in grass species.