Horticulture

Ideally, the internal levee should maintain a factor of safety of above 1.4, and as such, the maximum back flooding of the system will be will be MSL + 6 ft. Table 3 shows that for different river water levels increasing the back flood water levels results in a increase in the the factor of safety of the levees. While there are minor variations to this trend, the levee becomes more stable with increased back flooding. This result supports the theory that the hydraulic exit gradient resulting in piping and slope instability can be reduced by the Water Management System. The optimum back flooding levels from simulations of various elevated river conditions are highlighted in Table 3. Significantly higher factors of safety for the existing sections of the Sherman Island levees can be achieved with these levels of back flooding resulting in greater levee stability and lower risk of failure from river flooding. The driving forces for levee and piping resulting in slope failure stems from high exit gradients.As pressure builds up on one side of the levee, stacking flower pot tower water is forced towards the backside of the levee and thus increases the hydraulic gradient. This increases the already high hydraulic gradient of the peat layer even further.

The addition of the temporary flood storage zone allow for siphoning water to balance water levees on each side on each side of the level. Therefore the exit gradient and pore pressures are decreased leading to a significant increase of the stability of the levees. Figure 7 compares the exit gradient for an 8+MSL flood level and back flooding of 2+MSL which indicates a reduction of the exit gradient by approximately half.Site work estimates consider the construction costs of internal cutoff levee and necessary road improvements for site access. The cutoff levee requires approximately 440,000 cubic yards of fill at $15 per cubic yard, therefore the total cost of the levee is approximately $6.6 million . For project access, revitalization of a portion of the former Victory Highway is the most likely option. The approximate length of road to be improved is 0.54 miles with an estimated cost of $1.8 million per mile constructed . This value reflects construction equivalent to a one-foot asphalt concrete layer and gives a total of approximately $972,000. An alternative route involving revitalization of a larger portion of Victor Highway stretching roughly 2.17 miles would cost $3.9 million, which is four times the cost of the first option. The total materials cost for the aquaponics system is approximately $23 million. This includes 600 aquaponics subsystems at $30,000 to $50,000 each, with an average of approximately $38,500. The prices of hydroponics and aquaculture components, including: tanks, pumps, anchors, and rafts per subsystem.

The prices are based on data from UVI Aquaponics, CropKing Hydroponics, Dock Float Supply, Dock Builders Supply and Seaflex Anchors. Labor and construction costs are not included as further design and bidding by licensed aquaponics contractors are needed. The cost of wetland development is extremely variable depending on the scale of the project and quality of labor and materials used. According to Dr. Alexander Horne, estimates for wetland construction range from $2,000 to $75,000 dollars per acre . For the purposes of this project, Dr. Horne recommends a value of $25,000 per acre of wetland constructed. Therefore for 700 acres of wetland restoration there is an initial cost of approximately $17-18 million. Operations & Maintenance The economic analysis of sturgeon production by Logan estimates costs of operations and maintenance per aquaculture system at approximately $3,500 to $4,000. This includes cost of labor, feed, electricity, and medicine. Assuming a 25 percent increase in these costs to account for agricultural labor, this results in operating and maintenance costs between $4,000 and $5,000 per subsystem. Consequently for the 600 systems proposed there is an annual cost of approximately $2.8 million. For the proposed system on Sherman Island, there are three chief sources of revenue. The first is the sale of agricultural yields from the hydroponics component. The second is the sale of fish yields from the aquaculture component. The third source is the economic value placed on various wetland services.

Income generated by agriculture is dependent upon the type of crop produced as well as the frequency of harvesting cycles. For the purposes of this model, it is assumed that average yield is approximately 11,000 pounds of crop per year per system. Price indices for crops typically used in aquaponic operations indicate that cilantro has the highest economic benefit of $88,000 per year, with onion having the lowest value of $1,100 per year. On average, each year of production will generate roughly $25,000 to $30,000 in revenue per subsystem, resulting in nearly $17 million of income over a two-year time period. The aquaculture component of this business model uses price indices developed for similar systems rearing sturgeon. Numbers generated from previous examples include considerations for 5, 10, and 15 brood stocks. There is a harvest time of 18 months for each system, and production will likely be staggered to produce maximum output and more continuous income for the project. Considering a time period of two years, fish production gives estimated revenue of roughly $17 million assuming a sale value of $4.5 per pound of sturgeon. Wetland value is based on a sum of averages of the following services: gas regulation, disturbance regulation, water regulation, water supply, waste treatment, habitat, food production, raw materials, recreation, and cultural ecosystem services . Revenue for these services is presented in Table 4.The wetland component of the system will provide carbon sequestration. Carbon sequestration is the capture of atmospheric CO2 by plants such as bulrushes and cattails through photosynthesis. The Bulrush wetland can grow approximately 500 g C/ m2 per year, half of which is preserved. The CO2 is absorbed by the plants and then incorporated into the soil biomass as the plants decay over time. A wetland can sequester as much as 25 metric tons of CO2 per acre per year. This helps to reduce carbon emissions while reversing subsidence . The system actively stops and reverses subsidence. By flooding the area with water the peat layer preserved because it is wet. As the plant biomass from the wetland decomposes it increase the peat layer and therefore reverses subsidence. Wetland vegetation and non-sellable crop yields from the hydroponics system can be used to restore peat. Similar subsidence reversal projects have been proposed and approve on and around Sherman Island. One such project includes Twitchell Island Ponds where an increase of 1.6 inches/year in addition to preventing subsidence of approximately 1 inch/year has been demonstrated. During flooding even there will be sediment will be transported into the system and then settle out before the water in release back into river. Therefore sediment can slowly build up with flooding events and also reverse subsidence .

In addition the wetland restores natural habitats for birds and aquatic species. The fish a reptile species that would take advantage of this wetland ecosystem include carp, crappie, striped bass, white catfish, and pond turtles. Along with hundreds of different bird species including the mallard duck, red-tailed hawk, eagle, Canadian goose, cinnamon teal, ring-necked pheasant, ruddy duck, wood duck and many more. The wetlands will serve to restore the delta wetland habitats as well as to sustain future bio-diversity of wetland species . Water contamination and waste is reduced by the aquaponics system because nutrients and water are recycled back into the system through water recirculation. This separation of the system from the surrounding environment reduces water waste due to runoff and agricultural pollution when compared to terrestrial agriculture. Terrestrial agriculture requires a large amount of water,ebb and flow the majority of which is not absorbed by plants but leaves the land via groundwater runoff. Agricultural runoff promotes algal blooms and eutrophication, which lower the oxygen concentration dissolved in the water and leads to death of aquatic species. The potential environmental concerns of the wetland include the production of methane from biomass decomposition. While wetlands sequester carbon and hence reduce carbon dioxide, they release other GHGs, primarily methane and nitrous oxide . The concern is that Methane has a Global warming potential 21 times that of CO2, and N2O has a GWP 310 times that of CO2. Most of the GWP of CH4 occurs in the first twelve to fifteen years after its release, whereas the GWP of CO2 lasts decadnto the water within the system and the transport of such toxins into the surrounding environment. The soil at Sherman Island has been exposed to pesticides, fertilizers, and dredged soils for the bay which were contaminated with heavy metals. Therefore because the proposed design is to partially flood the soil the toxins could leach from the soil. In order to mitigate this possible issue a test flood plot an monitor the toxin release levels. Leach test or Barrel tests can be used to mimic the conditions and will give results that can be used to assess further needs of mitigation. Additionally, the heavy metals such as the metals off of ships including copper, zinc, and nickel can be immobilized in wet peat soils.

Organic compounds will bind to peat and once the wetlands have been established the plants will also remove toxins from the water. Finally because the flood storage zone will contain a large volume of water the toxins will be diluted. There may be short term effects from pesticides so to prevent birds from being exposed to these toxins the use of bird scaring devices can be implemented until toxins diffuse. It is important to monitor the toxins before, during and after the installation of the wetland in order to understand the mitigation strategies that must be considered and implemented. Many of the environmental concerns can be monitored at demonstration sites including Mayberry Farms Subsidence Reversal and Carbon Sequestration Project and the Twitchell Island Ponds which both contain many of the same environmental considerations that will be faced in the flood storage zone. Fatty acid composition was determined in leaves of purslane plants exposed to saline treatments for 30 days. A sample of 50 g of fresh tissue leaves was dehydrated in a incubator at 60 ºC for 48 h. The dehydrated tissue was macerated and stored in 15 ml vials at room temperature, protected from light. An aliquot of 1 g of each sample was used to determine fatty acid composition, according with the procedures described by Liu, et al. . After hexane extraction, the samples were vortex-mixed, centrifuged, and the upper phase was collected prior to GC analysis. Samples were injected via an auto sampler onto a fused-silica capillary column in a HP 6890 gas chromatograph system fitted with a flame ionization detector and eluted with helium at 44.0 ± 1 ml/min, with a split ratio of 1:17. The injector and detector were heated to 250 ºC. The column was temperature programmed from 130 ºC to 180 ºC at 25 ºC/min, and then to 230 ºC at 2.5 ºC/min. Fatty acid methyl esters were identified by comparing GC retention times with those of a mixture of standard fatty acids methyl esters Mix C14-C22, . Fatty acids were quantified using peak areas integration against internal standard. In GL and GR cultivars, submitted to 30 days of saline treatment, the main fatty acids detected were C16:0, C18:3 and C18:2. Such high amounts of palmitic acid, has it was found in the P. oleracea leaves of control plants is unusual, previous studies reported an proportion of 17% of this fatty acid in the total FAMEs detected . Palmitic acid, the most abundant fatty acid in the human diet, causes oxidative DNA damage, DNA strand breakage, necrosis and apoptosis in human cells in vitro, but when consumed with others fatty acids, like PUFAs, is unlikely to have any significant impact on human health. The amount of unsaturated fatty acids was also a little bit lower than expected, although the proportions between UFAs and SFAs were maintained ; it has been reported in other studies that the linolenic acid content in P. oleracea leaves was circa 50% of total fatty acids and linoleic acid content 3- 4 times lower.

However, only photosynthesis rates during leaf senescence of plant pre-cultured at low N supply reflected leaf senescence score during reproductive growth and N efficiency in the field experiments. Therefore, cultivar differences in leaf senescence during reproductive growth can only partly be reproduced in a short-term nutrient-solution experiment. Several differences between vegetative and reproductive growth might influence the induction and development of leaf senescence: first, although leaf senescence might be induced by N shortage both in hydroponics and under field conditions, the timing of N shortage is dependent upon different factors. In the field, the exploration of N sources in deeper soil layers might play the most important role for N uptake during reproductive growth . Thus in the field, root growth and morphology are the most important plant traits, which play only a minor role for N uptake in hydroponics. Secondly, source-sink relationships differ distinctly between vegetative and reproductive growth, both for carbohydrates and as a consequence also for N. The changes in assimilate flows might influence the development of leaf senescence, or at least the parameters used to characterize leaf senescence. However,potted blueberries the fact that photosynthesis rate during late stages of leaf senescence was significantly correlated to leaf senescence in the field experiments and to grain yield at limiting N supply suggests that cultivar differences in specific steps of leaf senescence related to the breakdown of the photosynthetic apparatus contribute to N efficiency in the field.

Life support systems are what make human travel a possibility. In long range space travels, such as the travel to Mars, life support cannot depend upon storage alone, it requires a fully regenerative system as well, i.e. waste must be reclaimed for reuse. Steam reformation, supercritical water oxidation, electrochemical oxidation, and incineration are a few of the solid waste reclamation technologies that are being developed and tested for use in space travel. Currently though, it seems that incineration might be the best choice among the previously mentioned, in providing a fully regenerative system. Through rapid conversion, incineration of the inedible parts of wastes and crops produces carbon dioxide, water, and minerals. Incineration is already the most thoroughly developed technology for use in a terrestrial environment. However, with the use of incineration in a closed environment, there is the eventual buildup of pollutants that are emitted in the process. The resulting NOx and SO2 pollutants need to be removed and recovered for reuse by a flue gas cleanup system. Important things to consider when developing a flue gas clean up technology for use in long range space travels are safety, energy requirements, sustainability, and doable under a micro-gravity condition. Due to the sensitivity and restrictions of space missions, a flue gas clean up system lacking in any of these considerations can be hazardous and could potentially compromise the missions. Technologies requiring things such as expendables or the use of catalysts are unsuitable for space missions due to the loss of valuable resources and the possibility of catalyst poisoning thus limiting the life-span of a catalyst. Also, due to the micro-gravity, it is difficult to use wet processes that handle liquids, such as spray absorbers.

Consequently, even though there are numerous flue gas clean up technologies developed , taking into consideration the limitation each provides, the number of reliable and applicable systems seem to be dwindling. Commercial activated carbon, made mostly from materials such as coconut shells and coal, has been studied for the adsorption and/or reduction of NOx and SO2 . In this paper, we study the use of the activated carbon prepared from hydroponic grown wheat straw and sweet potato stem for the control of air pollutants that are a result of incineration during space travel. Both wheat straw and sweet potato stalk are inedible biomass that can be continuously produced in the space vehicle. It was found that there is actually a minuscule amount of SO2 in the flue gas from the incineration of hydroponic biomass, and that most of the sulfur from the biomass ends up as sulfate in flyash. Since SO2 should have already reacted with the alkali metal, the technique entails the carbonization of the wheat straw and sweet potato stalk, resulting in an activated carbon for the adsorption of NOx and then a reduction of the adsorbed NOx by carbon to form N2. Since most NOx in flue gas from combustion is in the form of NO, and NO2 is readily adsorbed on the activated carbon, this paper concentrates on the removal of NO. Parametric studies on the production of activated carbon and the adsorption of NO by the carbon have been conducted. The optimal conditions and effectiveness of this procedure to regulate NO emissions have been determined.After the biomass was shredded, it was packed tightly into a stainless cylinder for pyrolysis and activation. Nitrogen and Carbon dioxide were used as the pyrolysis and activation gases, respectively. The gas flow rate for pyrolysis was 0.5-1.0L nitrogen per minute and activation was 0.25L carbon dioxide per minute.

The amount of biomass used was approximately 50.0-60.0g for each batch. Pyrolysis and activation temperature and times were varied during carbonization in order to determine optimal activation conditions. The activation temperature was 50°C higher than the pyrolysis temperature. The notation of the activated carbon “WS-2-600-1-650” and “SP-2-600-1-650” indicate that the activated carbon was prepared from wheat straw and sweet potato stalk, respectively with a 2 hrs pyrolysis time at 600°C followed by 1 hr of activation at 650°C. Once activation was complete, CO2 was supplied to the sample until it could be sealed in a container. Biomass on a space mission would likely come from a hydroponic system. The total mineral content of hydroponic biomass can be much higher than field-grown biomass if nutrients are supplied luxuriously to the hydroponic solution. A typical hydroponic plant solution has a nutrient ratio of N:K:Ca:P:S:Mg =16:6:4:2:1:1 and contains micro-nutrients of B, Mn, Zn, Cu, Co, Mo, and Fe. The concentration of potassium in the solution is about 6 mM. Thus, hydroponic activated carbon adsorption efficiencies may differ significantly from field-grown biomass. Field grown biomass generally has less than 10% mineral content, while hydroponic biomass can have as much as 30% mineral content. Because of this, many of the experiments were performed using char, which had been soaked in deionized water to remove the soluble minerals. After soaking, the char was drained and dried at 200o C for one hour before use. Specific surface areas of samples were determined by gas adsorption. An automated adsorption apparatus was employed for these measurements. Nitrogen adsorption/desorption was measured isothermally at -196o C. Before any such analysis, the sample was degassed at 250o C in a vacuum at about 10-3 torr. The nitrogen isotherms were analyzed by the BET equation, to determine the surface areas of the chars. BJH adsorption cumulative surface areas of pores of the samples also were determined.Most of NOx in flue gas from combustion is in the form of NO. Also, NO2 is readily adsorbed on the activated carbon. Consequently, efforts were directed to determine conditions for maximal removal efficiency of NO. Adsorption efficiency of NO on the activated carbon was studied to evaluate the effects of temperature, oxygen composition,square plastic pot moisture and flow rate on the production of effective activated carbon. The adsorption experiments were performed by using a simulate flue gas with variable concentrations of N2, carbon dioxide, oxygen, NO and H2O. NO and NO2 concentrations were analyzed by a chemilum inescent NOx analyzer . The amount of NOx absorbed by the activated carbon was determined from the difference in NOx concentration of the inlet and outlet gases. It was assumed that missing NOx was adsorbed by the activated carbon.Preparation of activated carbon was conducted by heating hydroponic grown wheat straw and sweet potato stalk under anaerobic conditions. Nitrogen and carbon dioxide were used as the pyrolysis and activating gases, respectively. Time and temperature were varied during pyrolysis and activation to determine optimal carbonization conditions. In order to determine optimal carbonization temperatures, samples of wheat straw and sweet potato stalk were heated at 100o C intervals between 300 to 800o C for two hours during pyrolysis and between 350°C to 850o C for one hour during activation. Afterwards, percent burn off was measured. Higher carbonization temperatures caused larger portions of the samples to burn off, as shown in Fig.2.

The burn off was 68% at 600°C and 86% at 800°C for wheat straw, while 66% at 600°C and 92%at 800°C for sweet potato stalk. Both samples appear to follow the same burn off trend, the percent burn off increases only slightly with the increase of temperature between 400°C to 600o C. However, pyrolysis and activation temperatures above 700o C were observed to cause significant amounts of wheat straw and sweet potato stalk to burn to ash. It is thus not recommended that reactions be run at temperatures exceeding 600o C. The decrease in surface area beyond 600°C is caused by sample burning off. Using temperatures much lower than 600o C would compromise the maximum amount of effective adsorption surface area attainable . It is not only important to run reactions at temperatures low enough to prevent burn off and ash formation but also high enough to generate effective surface areas, which would be at 600o C for wheat straw. Fig.4 shows the BJH adsorption cumulative pore area of wheat straw activated carbon generated at different pyrolysis and activation times. It is evident from the plot that samples derived from longer pyrolysis and carbonization times exhibited a higher micropore count compared to shorter times. As with temperature generation optimum, however, too long pyrolysis and activation reaction times cause an adverse increase in burn off percentage. Though wheat straw carbonized with a pyrolysis time of six hours and activation time of 2 hours still demonstrated a higher cumulative pore area than the shorter times, it also produces larger amounts of burn off. Since reaction temperatures were kept below 600o C, ash formation that would have diluted effective surface area was prevented. Even though ash does not form, pyrolysis and activation times must still be chosen to create a balance between pore formation and burn off, one that would generate a high micro-pore count but at the same time, minimize material loss. The optimal pyrolysis and activation times for wheat straw are 2 hours and 1 hour respectively.Optimal pyrolysis and activation temperatures and times for carbon preparation were determined based on the amount of NOx that can be adsorbed by the activated carbon. The adsorption capacities of wheat straw activated carbons generated by different pyrolysis and activation temperatures are shown in Fig.5. A gas mixture containing 250ppm of NO, 5% O2, 10% CO2, with N2 as the balance was passed, at a flow rate of 250ml/min, through a turbular reactor containing 2g of activation carbon at 25o C. It is evident from the plots that the WS-2-600-1-650 activated carbon had the best adsorption efficiency. Samples carbonized above 600o C have higher ash concentrations than those carbonized below, while those carbonized below have lower percent micro-pore counts and surface area—both explaining why wheat straw generated at 600o C had the best adsorption efficiency. The NO adsorption efficiencies of wheat straw samples carbonized by differing pyrolysis and activation times are shown in Figure 6. It is evident from the plots that activated carbons carbonized by prolonged pyrolysis and activation times have better adsorption efficiencies than those carbonized by shorter times due to higher pore count and BET surface area. The micro-pore count and the surface area of activated carbon increases with an increase of the preparation time, which explains why the samples with the longest pyrolysis and activation times have the best adsorption efficiencies. However, prolong activation results in more burn off and the production of ash. A balance must be reached when setting reaction parameters, one that will generate the largest surface area without a significant burn off. We have found that the optimal pyrolysis and activation times for wheat straw are two and one hour, respectively. The hydroponic biomass possesses high mineral content. The effect of the minerals on the activated carbon on NO adsorption efficiency was studied. The activated carbon was first soaked in water to dissolve the soluble minerals and then dried to remove the moisture from the carbon particles. The adsorption experiments using the mineral-free activated carbon were performed and the results indicate that the NO adsorption efficiency was substantially improved .

We refer to a farming system as “diversified” when it intentionally includes functional biodiversity at multiple spatial and/or temporal scales, through practices developed via traditional and/or agroecological scientific knowledge. Farmers manage this functional biodiversity to generate critical ecosystem services to agriculture . At the plot scale, diversified farming systems may include multiple genetic varieties of a given crop and/or multiple crops grown together as polycultures, and may stimulate biodiversity within the soil through addition of compost or manure . By crops, we mean either annual or perennial crops, including tree crops. At the field scale, DFS may include polycultures, non-crop plantings such as insectary strips, integration of livestock or fish with crops , and/or rotation of crops or livestock over time, including cover cropping and rotational grazing. Around the field, DFS may incorporate non-crop plantings on field borders such as living fences and hedgerows. At the landscape scale, DFS may include natural or semi-natural communities of plants and animals within the cropped landscape/region,4x8ft rolling benches such as fallow fields, riparian buffers, pastures, meadows, woodlots, ponds, marshes, streams, rivers, and lakes, or combinations thereof . The resulting heterogeneous landscapes support both desired components of biodiversity and “associated biodiversity”; together these two elements make up agrobiodiversity .

Components of the agrobiodiversity within DFS interact with one another and/or the physical environment to supply critical ecosystem services to the farming process, such as soil building, nitrogen fixation, nutrient cycling, water infiltration, pest or disease suppression, and pollination, thereby achieving a more sustainable form of agriculture that relies primarily upon inputs generated and regenerated within the agroecosystem, rather than primarily on external, often nonrenewable, inputs . Spatial considerations are important, since different components of the system must be in sufficient proximity, at each relevant scale, to create needed interactions and synergies. For example, the utility of intercropping for reducing below ground soil disease depends on spacing the different crops such that their root systems interact . Similarly, wild bee communities can only provide complete crop pollination services when a sufficient proportion of their natural habitat occurs within a given distance of crop fields . A DFS is not only spatially heterogeneous, but is variable across time, due both to human actions , and natural successional processes. Figure 1 presents the conceptual model of a DFS. The term agroecology goes back more than 80 years and originally referred to the ecological study of agricultural systems . Much agroecological work seeks to bring Western scientific knowledge into respectful dialogue with the local and indigenous knowledge that farmers use in managing ecological processes in existing agroecosystems . More recently this hybrid science has evolved to include the social and economic dimensions of food systems .

Partly in response to the industrialized agriculture of the Green Revolution , agroecology also came to mean the adoption of sustainable agricultural practices , and became an integral component of various social movements seeking alternatives to industrial agri-food systems. Thus agroecology currently holds multiple meanings, and can refer to an inter- or transdisciplinary science, a set of sustainable farming practices, and/or a social movement . DFS is not an alternative to agroecology. Rather, DFS is a framework that draws from agroecological, social, and conservation sciences to focus analytical and action-oriented attention toward farming systems in which cross-scale ecological diversification is a major mechanism for generating and regenerating ecosystem services and supplying critical inputs to farming. Agroecological principles and methods can be used to evaluate DFS and to design or revive processes of diversification . In this essay and series of articles, we explore the ramifications of DFS for both ecological health and socioeconomic welfare, as well as examining the intersection of DFS with existing industrialized agricultural systems, supply chains, and national and international policies. DFS are complex social-ecological systems that enable ecological diversification through the social institutions, practices, and governance processes that collectively manage food production and biodiversity . As many political ecology scholars emphasize, ecosystems are densely interconnected with social relationships . Ecological variables such as soil, water, and habitat help configure an array of farming practices, exchanges of food and resources, and landscape management decisions that, in turn, influence the structure and function of the ecosystem. Further, as ecosystem services are generated and regenerated within a DFS, the resulting social benefits in turn support the maintenance of the DFS, enhancing its ability to provision these services sustainably . This interplay underlies numerous historically occurring and emerging DFS worldwide. Conversely, socio-political and economic processes such as the decrease of access and control over seeds or increased dependence on commodity markets can intervene to disrupt such feedback cycles, thus weakening DFS. The industrialization of agriculture has led to growing homogeneity across food systems as farming techniques and markets become more standardized .

As a consequence, the complex social relationships underlying agriculture and ecosystem service provision have become less visible. Focusing on DFS can help farming communities, researchers, policy makers, and industry recognize and restore these relationships. At their core, DFS depend on agroecological principles that are developed in and through the social relationships among working farmers, their communities and environments, and researchers, including ecologists, anthropologists, agronomists, and ethnobiologists . As seen in the Kreman et al. examples these principles take varied forms depending on local conditions. To understand how DFS may develop, function, and evolve over time and space, the particular context of each DFS needs to be studied, paying particular attention to the politics and power relations that reciprocally shape its ecological conditions. Many DFS were developed through traditional and indigenous farming knowledge and agrobiodiversity that was accumulated over millennia . More recently, other DFS have been created through targeted agroecological studies designed by scientists to solve particular problems . Historically, much knowledge about biologically diverse farming practices has been created and shared through peer-to-peer learning within traditional farming communities and, more recently, also through their collaboration with researchers interested in further developing agroecology . These relationships continue to be critical to the growth of DFS in new societal contexts and geographic locations. Since the 1980s, with the rise of the Campesino-a-Campesino and La Via Campesina movements, institutions such as government agencies, domestic and international NGOs, and universities have become increasingly active in promoting and diffusing agroecological principles through research networks and programs . These actors have added new institutional dimensions to the social relationships that help sustain DFS. An illustration of how social and ecological systems interpenetrate within DFS is in the Andean highlands, where indigenous farmers have managed their lands agroecologically for 3,000 years . The ongoing interplay between human management and physical ecology has created a landscape of agroclimatic belts at different altitudes, each characterized by specific field rotation practices, terraces, and irrigation systems, and the selection of specific animals, crops, and crop varieties . Within these belts, traditional knowledge has helped sustain tremendous genetic diversity, by perpetuating adapted land races and wild relatives of crops. Social cooperation is essential to managing the verticality and heterogeneity of the Andean ecosystem. A barter economy based on reciprocity, for example, facilitated complementary exchanges of plants and animals between ecological zones along the steep elevation gradient . In industrialized systems in both developed and developing countries, farmers must now negotiate with corporate food buyers, buy agrochemical and seed inputs from agents, seek loans from bank officials, and work with agricultural extension experts trained in pesticide use. Farmers rely on such relationships to compete effectively in supply chains and to manage changing ecological conditions, such as pest outbreaks. Nonetheless, these particular types of relationships often push individual farms to increased dependence on banks, damaging livelihoods, and undermining collaborative social learning groups as farmers specialize in a single crop and maximize short-term yields through the use of external inputs, to meet loan repayments.

The economic pressures in these tightly linked systems generally corrode ecosystem services, which are the very foundation of support for potential DFS. Farmers in industrialized systems may also engage in exploitative relations with immigrant or impoverished laborers, paying inadequate wages and enforcing long hours,flood and drain table helping perpetuate the apparent cheapness of food. Industrial production creates a number of “distances” between producers and consumers such that information flow diminishes across the supply chain . Thus within the industrial agri-food system, consumers remain relatively ignorant about the conditions of production, and would be less able to choose between products based on sustainability criteria, if they value these, and to exercise their buying power in favor of DFS. In turn, the risk perceptions of consumers and corporations may inhibit the growth of DFS. For example, during the recent food safety scare in fresh leafy vegetables in California, corporate buyers insisted that growers remove native vegetation bordering fields that might attract wildlife. This action was taken largely to assuage consumer concerns, despite the lack of scientific support . In alternative agricultural systems such as organic or low-input farming, farmers can build particular forms of relationships that help sustain ecosystem services and social infrastructure more effectively. We discuss many of these relationships, including direct marketing, fair trade certification, and food justice movements. In developing and studying these alternative systems, however, researchers, policy makers, and NGOs often neglect race, socioeconomic, and gender issues, or sublimate them into a broad social justice category. Finding ways to be far more inclusive of diverse racial, gender, and socioeconomic groups can help strengthen the socialecological basis of agriculture. For instance, African-American growers once represented a sizable proportion of the U. S. farmer population, or one million in 1910, declining to 18,400 by 1997, due to race discrimination and violence, lack of land tenure , and multiple waves of economic migration from the South to urban centers . Many of these black farmers used DFS practices; their displacement helped create an opening for industrialized monocultures. Now, many new farmers in rural and urban areas are black, Latino, or Asian; there is evidence that these farmers are more likely than their established peers to embrace sustainable agriculture practices if adequately supported . Immigrants such as the Hmong may sometimes develop culturally relevant, more diversified food production enclaves within industrialized systems that preserve their traditions and provide livelihoods.They are developing new linkages between cities and nearby rural areas, potentially helping recreate DFS. For example, Will Allen founded Growing Power, an urban farming NGO that serves disadvantaged neighborhoods in Milwaukee and Chicago, attempting to encourage youth of all races to take up diversified farming. In Chicago, black activists and physicians have formed the Healthy Food Hub, a food aggregation NGO which sources produce from a historically black farming community, Pembroke Township, about an hour from Chicago. These efforts show how people can demand greater political agency in building a democratic DFS . New quantitative and qualitative research is badly needed to evaluate and critique the social benefits that DFS may provide in contrast to industrialized systems. In general, further analysis is needed to understand how the social elements of DFS can help generate and regenerate ecosystem services, thus maintaining diversified farming systems. In turn, more research is required on the political and socioeconomic interventions that could help rebuild or sustain the socialecological cycles that underlie DFS.DFS are often embedded in social, political, and economic conditions that differ from those accompanying industrialized monocultures , particularly with respect to core stakeholders, markets, and distribution systems. Yet, DFS may not always be able to realize their potential social-ecological benefits due to the lack of enabling environments. We explore how alternative agri-food networks and social movements relate to DFS and assess their potential to both maximize social benefits and promote DFS through their demands for food sovereignty and food justice. The agri-food systems approach reveals the interconnected systems of inputs, labor, land, capital, governance and knowledge that maintain specific types of agricultural production, distribution, and consumption systems . The governance and structure of the food system upstream from the farm, such as international agricultural trade liberalization policies that promote cheap food imports from industrial into developing countries, government subsidies for fossil fuel-based agrochemicals and commodity crops and irrigation projects that primarily benefit larger landholders , all help to maintain the industrialized agri-food system .

Coupling with landscape microclimate models provides not only the vertical inputs used by agricultural models, but also gradients along the landscape. Coupling with hydrological models provides water flow paths such as surface runoff, vertical and lateral groundwater flow, and interactions between shallow soil and groundwater zones and with adjacent surface water bodies . Water quality models provide sediment and solute transport along the landscape controlled by water flows, and other effects such as wind erosion. Integration and upscaling of landscapes into the watershed scale requires three-dimensional coupling of the surface and subsurface water, energy and mass transfers. Condon and Maxwell and Maxwell et al. provide more details on coupled versus integrated models. At this scale, the groundwater aquifer system typically transcends the boundaries of the watershed, necessitating regional scale analysis to evaluate not only the impacts of cropping and animal production systems on water quantity and quality, but also feed backs from the hydrological system into the agricultural system . Further, meso-scale rainfall and evapotranspiration distribution models control the local surface and subsurface flow intensities, pollution and abatement . At this scale, human effects through land use changes, and ecological dynamics and transitions on natural or protected lands are also important components needed to evaluate overall sustainability of agricultural systems .Although some efforts have gone into integrating biophysical models ,mobile vertical rack more is needed to enable comprehensive assessments of agricultural systems across scales and adequately address environmental and economic responses to decisions and policies.

The need to address decisions and policies at scales arises frequently in agricultural system modeling. Resolving the time and space scale differences among model components is often a major issue, particularly when component models are developed independently for different purposes. This problem arises, for example, when one attempts to create a model that combines crop and hydrology models, crop and economic models, or crop and climate models . There have also been efforts, starting in the early 2000s, in which dynamic models have been developed to provide forecasts over aggregated areas . Traditionally, climate model output for a grid cell is down scaled to produce weather data time series for points that are then fed into crop models. However, the land surface also influences climate; processes within the atmosphere and oceans, and on the land, are coupled and dynamically interact over space on timescales from fractions of seconds to thousands of years. Crops are a major component of the land surface of the globe, occupying about a quarter of all land area. Regional climate can be sensitive to large-scale changes in cropped areas that can result from changes in economic or climate conditions . Therefore, another direction for agricultural impact assessments at a large-scale is to dynamically couple crop simulation with models of land and atmospheric processes.Osborne et al., 2009 showed that, in some parts of the world, the impact of changes in cropped area on regional surface temperature can be of the same magnitude as regional human-induced climate change.

This result raises the question of whether or not new fully-coupled climate change impacts studies will revise our previous estimates of food security impacts. It is clear that the full coupling of crop simulations within global climate models is opening up new possibilities for studies of the impact of climate change on agricultural production – studies that capture some of the complex and important feed backs within the Earth system at a large scale. Limitations in the skill of large-area modeling of crop production and yield is dominated by the density of data used in the simulation. More data should equate to better skill. However, the skill of large-scale modeling is determined by the smallest data set, whether this is the grid cell with the shortest run of observed yields, or the data grid with the largest resolution . We have seen recent increases in the resolution of climate input data and global grids of crop management and soil information. In this field of agricultural modeling, any future increase in data resolution should produce more skillful model simulations.We next discuss the state of current agricultural system science relative to its capabilities and limitations in providing information to assist a wide range of decision makers represented by the five Use Cases. Each Use Case contains a set of interactions between systems and users in a particular environment in a systems analysis. The Use Cases are for developing and developed country settings, demonstrating a range of needs for widely different applications at different scales and levels of intensification. Antle et al. indicated that these Use Cases need crop, livestock, and farming system models. The question that we address here is whether current agricultural system models, existing sources of data, and existing decision support systems are adequate for providing information needed for these Use Cases.Can existing crop, livestock, and farming system models, data, and ICT tools provide the information that Sizani needs to advise the small farmer? The short answer is “No”; there are currently no easily accessible and usable applications that would allow her to analyze the particular farmer’s situation. Or apps that can connect with models in the “cloud” to make runs needed for her to advise the farmer. Although there are models that partially meet her needs, and there are well documented examples of using models to develop insights on productivity enhancement strategies in the face of resource constraints and climate risk they have not been integrated or are not packaged for use by this type of non-expert user.

Models can, for example, simulate responses of crops to soil and weather conditions as well as water and nitrogen fertilizer input but do not generally simulate actual yield in production situations where, weeds, pests or diseases are not controlled. Two of the most serious limitations of many crop-soil models are their inabilities to accurately simulate soil infertility and their failure to represent losses associated with the wide range of pest, disease, and weed species that damage crops. In many intensive production systems, soil fertility, weeds, pests, and diseases are controlled so that responses in those situations can be represented by the costs of management inputs and the production responses to climate and water management. Typically, cropping system models simulate yields that are higher than actual yields in farmers’ fields, which are reduced due to poor management. In addition, fields are usually not homogeneous; for example, spacing between plants may vary considerably,whereas the models assume homogeneity. However, if pest and disease data are observed and available, these data can be input to some existing crop models to compute yield loss associated with specific pests and to diagnose the reasons for the gap between potential and actual yield . Keating and McCown have shown, however, that expert application of well adapted models can still lead to useful insights on many of the key constraints to productivity enhancement in small-holder situations. Generally, farming system models now in use have some capabilities needed to analyze this Use Case. However, most farming system models are not developed to be easily implemented by non-expert users nor for farms with characteristics different from those for which they were developed. An exception to this is the TOA-MD farming system model , although that model also needs reliable data from farm surveys to simulate a population of farms in contrast to a particular farm.It is impractical for Sizani to collect information on a particular farm, go back to her office and work with an analyst to evaluate options for the farmer. Instead, data are needed to describe a range of farming systems so that she could select the combination of biophysical, farming system, and household characteristics from available data. This would include information to allow her to tailor inputs to most closely match the conditions of specific farms. This includes climate, soil, management practice, labor and other inputs available for production and marketing of outputs, typical pest and disease pressures, availability and prices for farming inputs, and other farm, economic, and environmental information. Generally, sufficient data on the biophysical, environmental, and socio-economic conditions of each farm or for a range of farm typologies in the regions are not available. Although some data, such as climate and soil data, are available, generally these are not organized nor are they sufficiently site-specific that agricultural systems models can readily access them for analysis of specific farms. Although research has shown that some analyses needed to advise a farmer can be made, the availability of input data for agricultural systems models remains a major limitation.Most existing DSS tools that are available in Apps are focused on relatively narrow issues ,vertical grow rack such as when to apply a fungicide to a particular crop, when to apply the next irrigation, or how much N fertilizer to apply to a particular crop that will be grown on a particular type of soil in a specific setting. There are few DSS tools that make use of more integrated models to help advisors advise farmers in making farming system decisions . We envision a DSS platform that will connect various models, databases, analysis, and information synthesis tools in an easy-to-use interface for Sizani to set up the analyses and outputs to answer questions about the management of that particular farms’ biophysical and socioeconomic situation and the uncertainties in those estimates. Such DSS platforms are possible, but not yet constructed.Models of maize and other crops, livestock, and the farm household are also needed for this Use Case. These models are available for at least partially performing this type of analysis. Starting in the 1980s, several groups began using crop simulation models to evaluate alternative management systems in developing countries .

Models used in those efforts were generally based on CERES and other crop models now in DSSAT and on the ORYZA rice model developed by IRRI. More recently, the Global Futures and Harvest Choice CGIAR research projects led by the International Food Policy Research Institute have used crop and economic models to evaluate the potential benefits of developing new technologies, including new crop varieties . For example, Singh et al. used the DSSAT CROPGRO groundnut model with climate and soil inputs at six locations in India to evaluate different crop traits being targeted by CGIAR plant breeders. They found that the effect of combining various traits was beneficial, with estimated yield gains varying, depending on location and climate change conditions. Rapid advances in biotechnology and molecular plant breeding are helping researchers incorporate molecular markers and genes into models so that ultimately genetic composition of crops can be used to predict performance of future varieties to help target expensive and time consuming plant breeding efforts . The paper by Hwang et al. presents some concepts now being explored for next generation crop models. Similarly, considerable work has been done on farming system models to evaluate options for improving the livelihoods of farmers. These include farm simulation models , optimization models that attempt to select the best combination of enterprises and their management to achieve one or multiple goals of the farmer . Also, the Trade off Analysis model is currently being used as the basis for model-based impact assessments . Furthermore, this approach can incorporate results from crop and livestock models, as well as environmental and social outcome models, and it can be adapted for smallholder or large commercial farming systems. However, there are important limitations in the capabilities of these models, similar to those mentions in Use Case 1 Thus, there may be large yield gaps between actual yields in farmers’ fields and the potential productivity in those fields . When water, nitrogen, and climate are the major limitations in crop productivity, current models are highly useful, assuming that soil, weather, cultivar, and management input data are available for the analyses. In this Use Case, it is likely that other factors, including other soil nutrients, pests, diseases, and weeds, need to be taken into account. The challenge for next generation models includes not only modeling those factors but also collecting data that describe the production situation with all of the important yield-limiting and reducing factors. Another question is whether existing biophysical models can predict performance of the wide range of intensification options that may be used by farmers for this Use Case.

A variety of labels have been used for this emergent industry: precision agriculture, e-agriculture, smart agriculture, and digital agriculture, among others. Despite early critical use of precision agriculture, the term tends to be used in the industry to signify a specific suite of production-oriented technologies.However, information technologies are also used to open new markets and new territories for production. For example, digital platforms have become increasingly important for individual producers to bring their goods to market. Figure 1 shows how information technologies are intertwined throughout the cycle of agricultural production and sale.We use digital agriculture for its semantic breadth and increasing currency. In our taxonomy, precision agriculture is a subset of digital tools which improve efficiency through careful management of inputs. Three other types of tools—marketplace and financial platforms, e-extension, and smallholder management—are typically platform-based systems that mediate the social relation between farmers and the outside world. Marketplace and financial technologies help farmers access new credit lines and optimize their market behavior. E-extension is the digitalization of the practice of implementing technological innovations through farmer education, particularly in the international development context. E-extension, like the analog version that preceded it, dutch buckets system is largely reliant on insights produced far from the farm.

Finally, smallholder management platforms allow larger agribusinesses to exert control over smallholder farmers through close management of their inputs, products, and so forth. This may allow major actors to divest themselves of the risk inherent in owning land and instead subcontract smallholders in a relationship analogous to other platforms in the gig economy.For digital agriculture’s boosters, it has the potential to be the much-needed “fourth agricultural revolution” . In particular, it is framed as a climate-friendly way to feed the world and improve the lot of farmers around the world. By making the application of inputs more efficient, digital agriculture can indeed lessen the environmental impact and yield of agriculture. By increasing input efficiency and improving knowledge of market demand, digital agriculture may indeed improve the fortunes of producers. The rhetoric is not dishonest, but it is incomplete.Optimizing inputs enables the continued use of ecologically-harmful chemicals and practices, which would otherwise be abandoned if their effects were not actively mitigated . Digital agriculture’s marketing claims it will improve efficiency, increasing yield and minimizing the use of inputs—many of which are harmful and unsustainable. The externalities produced by using these inputs are the “un- and undervalued costs of industrial capitalist agriculture” .

A team at Cornell, for example, has developed a model that recommends ideal fertilizer application rates for each section of a farmer’s field in order to minimize nitrogen run of into the Gulf of Mexico, which causes algal blooms, depletes oxygen levels in the water, and kills fish and wildlife.While optimization limits the short-term damage of unsustainable practices, it also makes those practices more politically permissible and financially feasible. Thus, by making unsustainable practices appear sustainable, the necessity of adopting more ecologically and socially sustainable and just practices is delayed. By focusing on input management, these technologies advance a limited interpretation of sustainability that still depends on of-farm inputs, rather than a more radical shift to permanently sustainable practices . Just as digital agriculture promises to minimize inputs, it also promises to maximize yield—yet yield is not the problem. In the 1970s Amartya Sen noted that while starvation was increasing globally, food per capital was also increasing —as population grew, food production grew at a greater rate, not only globally but even regionally. While some scholars have taken issue with Sen’s empirical basis, an updated analysis using 2010 statistics found the same results . The direct relationship between hunger and food per capital, when we would expect an inverted one, betrays the simple thesis that hunger is due to a lack of food availability. Instead, Sen attributes hunger to an inability to exchange for food. Davis similarly notes the disconnect between food availability and hunger, finding that famine can occur in areas of grain surplus because it is more attributable to rural food management and exploitation than to production .

The “solution” to hunger, then, lies not in yield. Yield has increased; food per capital has increased; hunger persists. Therefore, stretching yield through digital agriculture is insufficient and does not address the political-economic basis of systemic hunger.The third key claim made by digital agriculture’s boosters is that it will improve farmers’ welfare, in particular their profits. Profits may be found in better decision-making, better yields, and better access to market information . In the Global North, such increased profits may be plausible. However, a primary mode for digital agriculture, the platform service, means that the data produced typically becomes the property of the platform provider. Weersink et al. note that a key challenge for digital agriculture is making this data useful; this, in turn, may favor larger companies with the capacity to process the data. Bronson notes this dynamic and warns that it may reproduce the distributional effects of the Green Revolution—that is, to concentrate wealth and power in the hands of major agribusinesses. In the Global South, digital agriculture presents a different set of problems for farmers’ welfare. Technological innovation that increases a crop’s yield in turn increases supply and undercuts the socially necessary labor time required to produce it. This dynamic lowers the crop’s exchange value at the expense of those at the bottom of global commodity chains, in particular the growers’ compensation per unit of crop. As this price drop is not accompanied by any increase in production for farmers without access to this technological innovation, this drop translates to lower overall compensation and to “exchange entitlement decline” . If they depend on exchange for subsistence, the decreased compensation translates to hunger as well.In reflecting on these mainstream claims, a different theme emerges. Rather than sustainability, nourishment, or farmer welfare, digital agriculture is fundamentally about securing the conditions to generate profit in the food system. Crucially, however, this is not about profit in food production alone, but in the wider capitalist economy for which food is obviously a fundamental input. Therefore, we submit that digital agriculture must be understood as addressing a specific set of crisis tendencies that have emerged at a particular juncture in the social, ecological, and spatial history of capitalism. This juncture is defined by interlocking moments of ecological disaster; enormous advances in information production, gathering, and processing; and “hypertrophic” urbanization . In this section we argue that rather than a solution to the climate crisis, hunger, or farmer welfare, the rise of digital agriculture can better be understood as an attempt to overcome crisis tendencies of “the relentless growth imperatives of an accelerating, increasingly planetary formation of capitalist urbanization” .

After briefly excavating the informational dynamics latent within the framework of extended and concentrated urbanization, we describe how digital agriculture functions as a “data fix” by allowing the intensification of agricultural industrialization and the extraction and enclosure, for eventual profit, of the data produced by digital agriculture technologies. An early theme in globalization literature was a tendency to embrace the rise of information technologies in a way that dematerialized the now planetary systems of extraction, production, and consumption . Such concepts, however, have largely been absorbed by analyses which show that a deterritorialized “information society” is not displacing traditional modes of production and social relations as much as emerging as a financial-managerial stratum in a “new international division of labor.” Another major theme in globalization studies is the ‘global city network,’ a set of nodes in the global space of flows from which the global economy could be commanded and controlled . In describing such cities as “strategic sites where global processes materialize” , they appear to be material sites floating in a sea of immaterial processes. In this model, cities are simultaneously the result of, yet alienated from, specific material processes— such as agricultural production—taking place beyond their bounds. In both concepts the informational nature of globalization is over-emphasized at the expense of its material effects. In an era of climate crisis, this shortcoming is glaring.One response has been to radically reframe globalization as a material process of urbanization,dutch buckets which unfolds as the product of dialectically-entwined moments of extension and concentration . Concentrated urbanization signifies the moment of agglomeration where the material flows of global capitalism accumulate into cities, megalopolises, and mega-regions. On the flip side, extended urbanization is the moment where remote territories are enclosed and transformed into operational landscapes that funnel energy, materials, and food into areas of accumulation. Both moments cause and are caused by the other: “The urban unfolds into the countryside just as the countryside folds back into the city” . Global capitalist urbanization is a metabolic process of moving and consuming the material world . This involves both fragmentation and homogenization —for example, the simultaneous expansion of monoculture agriculture and of liberal private property regimes. At the same time, enclosure and technological advances deprive peasants of their livelihoods; ‘depeasantization’ is the mirror of urbanization. However, the desire to develop a more materialist model of globalization leads to the black-boxing of information‘s role in facilitating vast networks of production and exchange. To bring information back in requires recognizing that something happens at the moment of concentration which sets the stage for extension. In the present framework, production and the growth imperative drive a search for more raw materials. But extension also depends on informational infrastructure to make a massively decentralized network of global supply chains profitable. Indeed, another way to describe capitalist geography is as “a skein of somewhat longer networks that rather inadequately embrace the world on the basis of points that become centers of calculation” . Information, along with material, is being drawn inwards in the moment of concentration; the processing of raw information—which is “what remains after one abstracts from the material aspects of physical reality” —into actionable knowledge informs extension processes. “Information processing” is computation, and computation at the scale required to make legible the vast amounts of data produced in the contemporary economy involves enormous physical infrastructural investment in data centers, undersea cables, and satellite networks . Such computational capital consists also of intellectual and human capital in the form of models, algorithms, and the expertise to deploy them.

There is a potential for the over accumulation of computational capital, however; as a result, there is a constant drive for firms to find productive outlets. This is what leads firms like Amazon, Microsoft, Google, Oracle, and Cisco—as well as funds invested in and consultancies hired by them—into digital agriculture. By locating, extracting, and enclosing data relevant to another materially productive sector , a firm like Amazon—whose cloud computing infrastructure Jef Bezos has compared to power utilities—can continue to grow. This applies at the worker level, too. Just as a glut of NASA-trained engineers and physicists became quants for hedge funds after the Space Race , a glut of software engineers and data scientists which Silicon Valley cannot absorb find employment outside of the tech sector, including at digital agriculture startups or divisions within larger agribusinesses. Indeed, agribusiness are planning for a future in which they become tech companies themselves: the head of digital agriculture at Bayer Monsanto, for example, has described the future of the conglomerate as a digital platform . The fundamental material crisis that digital agriculture attempts to fix through the manipulation of data is in the sociol-metabolic processes of capitalism and capitalist urbanization. To support social reproduction for a growing non-agrarian population, present-day industrial agriculture destroys its own ecological foundations. As the consequences of climate change become ever more apparent and render growing conditions ever more difficult, a new ecological regime is needed to prolong the production of cheap food and ensure future accumulation in the face of known crises . But not only is fossil fuel-based industrial agricultural production partially responsible for climate change—up to one-fifth of all greenhouse gas emissions—it also exhausts the ecologies within which it is practiced. The search for the fourth agricultural revolution is not a straightforward matter of addressing a Malthusian crisis of natural population growth, but a crisis of capitalism itself. This crisis tendency arises from capitalism’s dependency on the “four cheaps”— labor, food, energy, raw materials—to maintain each cycle of accumulation.

During the rest of the year, there are much less employment opportunities for rural than urban households, with the former working about half the time worked by urban households during the low season . Lack of labor smoothing across months can thus be a major cause of income differentials between rural and urban households. Measuring annual labor productivity as median household real consumption per capita, rural households are at 57% of individuals in urban households. When this is measured not per year but per hour worked, rural households are at 81% of individuals in urban households. With high urban unemployment in Malawi limiting the option of reducing rural poverty through permanent or seasonal rural-urban migration, this suggests that a key instrument for rural poverty reduction is to have less idle time for land and labor throughout the monthly calendar. For Bangladesh, Lagakos et al. proposed filling labor calendars for rural households through migration to cities during the lean season. When this option is not available due to high urban unemployment filling and smoothing labor calendars in rural areas becomes a key dimension of poverty reduction. This can involve employment both in agriculture with more diversified farming systems and in the local rural non-farm economy. This is the purpose of the agricultural and rural transformations that are important in redefining how to use agriculture for development.

Based on work done for the IFAD Rural Development Report led by Binswanger, for China by Huang , by BRAC on graduating the ultra-poor out of poverty ,nft hydroponic for the Gates Foundation by Boettiger et al. , and for the ATAI project , a strategy of using agriculture for development would involve the following five steps: Asset building, Green Revolution, Agricultural Transformation, Rural Transformation, and ultimately Structural Transformation as described in Table 1.Minimum asset endowments for SHF under the form of land, capital, health, knowledge and skills, and social capital are needed to initiate production for the market and participation in a value chain. This corresponds to minimum capital endowments to get started in production in farm household models such as Eswaran and Kotwal’s , and to asset thresholds to escape poverty traps in Barrett and Carter . The BRAC graduation model for the rural ultra-poor thus importantly starts with achieving minimum asset thresholds for households to engage in self-employment in agriculture , with rigorous impact evaluations demonstrating success in raising household consumption in five of six case countries. Evaluation with a randomized experiment of a BRAC credit program for landless workers and SHF in Bangladesh shows that loans can be used to achieve minimum asset endowments by renting land and selecting more favorable fixed rent over sharecropping contracts . The Green Revolution, whereby productivity growth is achieved in staple crops through the adoption and diffusion of high yielding variety seeds and fertilizers is the initial step in agricultural modernization. It has been actively pursued to achieve food security and is a learning ground for the subsequent transformations of agriculture and rural areas.

It has been a major success of the Consultative Group in International Agricultural Research and is still an ongoing effort in Sub-Saharan Africa and Eastern India. A key objective of the Agricultural Transformation is to fill in rural households’ labor calendars over as much of the year as possible through multiple cropping — which typically requires water control to cultivate land in the dry season–, the development of value chains for new crops, and contracting among agents in these value chains. An example is the introduction of short duration rice varieties in Bangladesh that frees the land for an additional crop, typically high value products such as potatoes and onions, between rainy season and dry season rice crops. This makes an important contribution to filling land and labor calendars and to reducing the length of the hungry season . Because the Agricultural Transformation implies diversification of farming systems, it is a key element of national food security strategies where diverse diets, including perishable goods such as fruits and vegetables, dairy products, and meats that are less traded than staple foods, are an important element of healthy diets . SHFs are engaged in value chains that define the way they relate to markets. Value chains for agricultural products link farmers backward to their input and technology suppliers and forward to intermediaries, processors, and ultimately consumers . Relations within value chains often take the form of contractual arrangements. Induced by income gains for consumers, urbanization, and globalization, there has been in recent years a rapid development of value chains not only for low-value staple food crops, but also for medium value traditional domestic consumption and export crops, and high-value non-traditional export crops. Their structure can take a wide variety of forms in linking SHF to consumers, ranging from traditional spot markets to elaborate contract farming, productive alliances , and out-grower schemes.Contracts can be “resource-providing”, thus contributing to solve market and institutional failures for participating SHFs.

A key objective of the Rural Transformation is to give access to smallholder households to sources of income beyond agriculture. In Ghana, income derived from the rural non-farm economy for rural households is about 40% of total income, a share that increases as land endowments fall . It is indeed the case that, with land limitations, smallholder households rarely exit poverty with agriculture alone. A rural transformation requires the development of land markets and of labor markets . This process will typically happen first in the more favorable areas where a rural non-farm economy linked to agriculture can develop through forward, backward, and final demand linkages. It corresponds to the Agriculture Demand-Led Industrialization strategy advocated by Adelman and Mellor that is actively pursued in countries such as Ethiopia and Rwanda, and through CAADP in much of Sub-Saharan Africa. There are basically two contrasted approaches to potentially overcoming the problems that obstruct an Agriculture for Development sequence. The first consists in focusing on particular groups of farmers and addressing each of the problems in their own shapes and forms that affect them in modernizing. We can label this a “supply-side” approach to modernization and transformations. It consists in securing the existence and profitability of innovations, ensuring their local availability, and then overcoming each of the four major constraints to demand and adoption through either better technology or through institutional innovations . The agents for this approach are principally public and social such as governments, development agencies, NGOs, and donors. The second consists in creating incentives for SHF to modernize by building value chains for the particular product, and managing vertical and horizontal coordination within the value chains to overcome the profitability-availability-constraints obstacles as they apply to inclusion and competitiveness of SHF in the value chain. This is a “demand-side” approach to modernization and transformations. It consists in creating the demand for innovations in order to establish SHF competitiveness within a value chain, and then securing the existence, availability, and conditions for adoption of innovations. The approach thus requires both value chain development and value chain inclusion of SHFs. In this case, the agents are principally private such as enterprises and producer organizations for contracting, and lead firms, multi-stakeholder platforms, and benevolent agents for coordination. Public-private partnerships can be found among both supply- and demand-side initiatives.Technological innovation are first analyzed in experimental plots, usually for yield and resilience to specific shocks. But this does not tell us whether the innovation is likely to be adopted by SHF. Analysis of the adoption problem should start with verification that the innovation is indeed profitable for the intended SHF under their own circumstances, objectives, and capacities. Measuring profitability in farmers’ plots is however very difficult . There are data problems in observing family labor time and definitional problems in establishing the opportunity cost for family labor and self-provided inputs. Conditions also vary year-to-year due to weather conditions, with only short time series to observe how climate affects outcomes, made even more difficult to interpret with climate change. And there are many unobservable conditions and complementary factors that affect profitability and compromise the external validity of any measurement made at a particular time and place. An alternative approach is to verify profitability without measuring it. Some among the best endowed and best located farmers have to be able to make sustained use of the innovation for the innovation to have adoption potential by others under current market, policy, nft system and complementary input conditions. This can be established by observation, experimentation, or simulation.

Once the innovation is proven profitable and is locally available, its adoption may still be hampered by constraints facing SHF in accessing liquidity, risk-reducing instruments, information, and markets. These four categories of constraints have been extensively analyzed using in particular randomized control trials to identify their causal relations to adoption . These studies typically seek to identify ways of overcoming these constraints that could be implemented by governments, international organizations, NGOs, and benevolent agents such as philanthropic foundations and corporate social responsibility initiatives. Due to seasonality, especially under rainfed farming conditions which is where most of the lag in modernization currently prevails , there is a lack of correspondence between the timing of agricultural incomes and that of expenditures. As a consequence, the inter-temporal displacement of liquidity through credit and savings appears to be important for farmers to invest in new technologies, purchase inputs, optimize the timing of sales, buy consumption goods, and cover timely expenditures such as school fees. Financial services for SHFs appear to frequently be ill-designed for their purpose, expensive, excessively risky, and not easily available. Even when they have formal land titles, SHF are typically unwilling to put their land at risk as collateral with a commercial bank, thus acting as “risk constrained” . Microfinance products that effectively circumvent the collateral problem by relying on group lending and joint liability tend to be too expensive for the long agricultural cycles and have repayment conditions that are typically ill adapted to the timing of farmers’ capacity to pay . Availability of credit from formal sources, both commercial and non-profit, is consequently limited, and SHFs must either self-finance or rely on informal lenders with prohibitive interest rates. Hence, there would appear to exist a largely unresolved liquidity constraint on adoption originating on the supply side of the financial market. Yet, this is often not the main reason for low adoption which may be on the demand side. Recent field experiments are providing evaluations of interventions aiming at relaxing the liquidity constraint on SHFs, with fertilizer the most commonly used indicator of technology adoption because of its ubiquitous recognition and yet massive under use. While contexts and interventions vary for these experiments, they surprisingly tend to show that a liquidity constraint is not the reason why a majority of SHFs are under-investing in fertilizers. The main constraint may be instead lack of profitability in adopting fertilizers.A first category of experiments consists in providing unrestricted access to credit to a defined eligible population, as was done in Morocco , Mali , and Ethiopia . While interest rates in these studies were variously subsidized , uptake remained low: only 17% of eligible farmers took a loan in Morocco, 21% in Mali, and 36% in Ethiopia. Furthermore, farmers that did take a loan only used a small fraction of the liquidity to increase their expenditures on fertilizer or other agricultural inputs . Other experiments offered restricted credit that can only be used to purchase agricultural inputs. Such credit displaces the equilibrium allocation of liquidity in favor of the targeted inputs, similarly to what a price discount would do. And yet, uptake remained low. In Malawi input credit for high-yielding maize and groundnuts was taken by 33% of the farmers . This low demand for credit thus seems to be reflective of a low demand for the inputs themselves. Low demand for fertilizer is exemplified in two rather extreme experiments. In Mali, Beaman et al. provided to another group of farmers a pure cash grant, rather than the credit described above. This only increased expenditures on fertilizer by 15%, in comparison with 11% with a credit that had to be paid for, showing that credit is not the major constraint to adoption.