Sustainability & Environment (Agri-Food Systems) · Sustainable Agriculture & Agroecology · also Farming Systems and Sustainable Agriculture, Integrated Farming Systems, Sustainable Intensification
The whole farm as a system: crop-livestock-fish-tree integration, rainfed and smallholder systems, sustainable intensification and how farm sustainability is measured. Builds on the agriculture library subject Farming Systems & Sustainable Agriculture and goes into assessment.
In the exams: ICAR AIEEA-PG Agronomy and Physical Science groups; ICAR AICE Agronomy (farming systems)
A farming system is the whole farm seen as a managed system: crops, animals, trees, fish, soil, water, labour, capital and the household's goals, linked by flows of materials, energy, money and information. Farming systems analysis asks how these parts fit together, what limits the system, and how changes in one part affect the rest. Sustainable and integrated farming systems apply this view to make farms more productive, efficient, resilient and fair over time.
The subject matters because most decisions that determine sustainability are taken at farm level and involve trade-offs between enterprises. A dairy herd that supplies manure to crops, a fish pond that recycles poultry litter, or a dryland farm that combines drought-tolerant crops with small ruminants can each perform better as a whole than their parts would separately, but integration also adds labour, risk and management complexity. Policy debates about sustainable intensification, natural farming and smallholder resilience all depend on how farming systems are measured and compared.
This manual is written for agronomy and environmental science students preparing for Indian postgraduate and PhD entrance tests, for MSc sustainable agriculture courses elsewhere, and for advisers and analysts who assess farms. It concentrates on whole-farm analysis: cropping system indices, integration, rainfed systems, sustainable intensification, resource-use and energy efficiency, sustainability indicators, livelihoods and the evidence on low-input systems. Ecological principles and transition frameworks are covered in the agroecology-principles subject, and tillage, residue and cover-crop practices in conservation-regenerative-agriculture; this manual refers to them rather than repeating them. Calculations are worked where examiners and analysts use them, and coefficients that vary between sources are labelled as such.
A system is a set of components that interact within a boundary to achieve some function. Applied to farms, the components are enterprises (crops, livestock, fish, trees, off-farm work), resources (land, water, labour, capital, knowledge) and the household that manages them. Key systems ideas are the boundary (what is inside and outside the system studied), inputs and outputs that cross it, internal flows between components, feedback loops that amplify or dampen change, hierarchy (field, farm, village, watershed, region) and emergent properties: properties of the whole, such as resilience or profitability, that cannot be seen in any single part.
Farms are open systems: they import seed, fertiliser, feed, energy and information and export products and wastes. They are also purposeful, because their managers have goals, and the goals differ: food security, cash income, risk reduction, status or keeping land in the family. Two farms with identical resources may be managed very differently because their goals differ, which is why farming systems research always studies the household as well as the field.
Farming systems research (FSR) developed in the 1970s and 1980s, largely in international research centres and national programmes, as a reaction to research that tested single technologies on research stations and then wondered why farmers did not adopt them. FSR starts with diagnosis of the farm system and its constraints, designs and tests improvements on farms with farmers, and evaluates them in the context of the whole system. Its participatory successors are discussed in the agroecology subject.
At larger scales, farming systems are classified to target research and policy. A widely cited FAO and World Bank study by Dixon and colleagues (2001) defined 72 broad farming systems across six developing regions, grouped by factors such as water source (irrigated or rainfed), dominant livelihood (crops, livestock, mixed, pastoral, fishing), market access and agro-ecology, such as the rice-wheat system of South Asia or the maize mixed system of East and Southern Africa. Such classifications help identify where a technology or policy is likely to fit.
In India, the ICAR Indian Institute of Farming Systems Research at Modipuram and the All India Coordinated Research Project on integrated farming systems develop and test farming system models for different agro-climatic zones, with special attention to marginal and small farms.
A systems view changes the questions asked. Instead of 'what is the best maize variety?', it asks 'how does a new variety change labour peaks, fodder supply, cash flow and risk across the year?'. Tools include resource-flow maps, seasonal calendars of labour and cash, enterprise budgets, whole-farm budgets and simple models. Leverage points are places where a small change has large effects, such as fodder supply in the dry season for a mixed farm, or water at a critical crop stage in a rainfed system.
Systems thinking also warns of unintended consequences. Subsidising one input may shift labour away from other enterprises; introducing high-yielding dairy cows may require purchased feed and expose the farm to price risk. Sustainability judgements therefore require looking at the whole farm and, often, beyond it to the watershed and the market.
Simple quantitative models help. A whole-farm budget adds the gross margins of all enterprises and subtracts fixed costs; a linear programming model chooses the enterprise mix that maximises income subject to land, labour, water and capital limits. Even without formal models, listing each enterprise's demands on scarce resources month by month often shows why a technically attractive change is not adopted.
A cropping system is the cropping pattern of a farm and its interaction with farm resources, other enterprises and technology. A cropping pattern is the yearly sequence and spatial arrangement of crops on a given area. Types include monocropping (one crop each year), multiple cropping (two or more crops on the same field in a year), which may be sequential (one after another), relay (the second crop sown into the standing first crop before its harvest) or ratoon (regrowth from stubble, as in sugarcane), and intercropping (two or more crops at the same time; see agroecology-principles for the land equivalent ratio).
Major cropping systems of South Asia include rice-wheat in the Indo-Gangetic Plains (on the order of 13 million hectares across the region), rice-rice in the humid south and east, cotton-wheat, sugarcane-based systems and rainfed systems based on sorghum, pearl millet, pulses and oilseeds. Each has characteristic sustainability issues: groundwater depletion and residue burning in rice-wheat, salinity in irrigated cotton-wheat, and drought risk in rainfed systems.
Cropping intensity is the gross cropped area divided by the net sown area, times 100: a farm with 120 ha net sown and 180 ha gross cropped has a cropping intensity of 150 percent. The multiple cropping index is the same idea applied to a region. Land use efficiency expresses the share of the year that land is occupied by crops: (sum of crop durations / 365) x 100; rice of 120 days and wheat of 140 days give 260 / 365 = 71 percent.
To compare systems with different crops, yields are converted to a common equivalent. Rice equivalent yield (or any base-crop equivalent) = sum over crops of (yield of crop i x price of crop i / price of the base crop). With rice at 5 t/ha and wheat at 4 t/ha, and prices of 20 per kg for rice and 22 per kg for wheat, the rice equivalent yield is 5 + 4 x 22 / 20 = 9.4 t/ha. Production efficiency is total output per day of field occupation: 9,000 kg from 260 crop-days is about 35 kg/ha/day. For intercrops, the area-time equivalency ratio (ATER) refines LER by weighting each crop's relative yield by its duration. These indices answer different questions, so analysts state which one they use and why.
Indices are only as good as their inputs. Prices change between years, so equivalent yields shift with markets, and they say nothing about water use, soil health or labour; they are usually reported alongside economic and environmental measures.
Diversified cropping systems include legumes for biological nitrogen fixation, crops with different rooting depths, and crops that break pest and disease cycles. In the rice-wheat system, options include replacing part of the rice area with maize, pulses or oilseeds, adding a short mungbean crop in the summer fallow, direct-seeded rice and zero-till wheat. In rainfed systems, intercropping of cereals with pulses (such as sorghum with pigeonpea) spreads risk and makes use of different growing periods. Detailed rotation effects and cover crops are covered in conservation-regenerative-agriculture; this chapter's concern is how to measure and compare cropping systems at farm level.
Good design also considers timeliness. In intensive systems, the turnaround time between harvest of one crop and sowing of the next decides whether a third crop fits, and late sowing often costs more yield than any input saving. Short-duration varieties, zero-till sowing and mechanised harvesting are therefore as important to cropping intensity as crop choice. In rainfed systems, matching crop duration to the length of the growing period, so that the crop matures before soil moisture runs out, is the first design rule; long-duration crops on shallow soils frequently fail in years of early monsoon withdrawal.
An integrated farming system (IFS) links enterprises so that the by-products of one become inputs for another: crop residues feed animals, manure fertilises crops and fish ponds, pond silt returns to fields, and trees supply fodder, fuel and shade. The aims are to recycle nutrients and energy within the farm, reduce purchased inputs, use family labour more evenly over the year, spread risk across enterprises and raise income per unit of land. These synergies are an application of the diversity and recycling principles discussed in agroecology-principles.
Integration has costs and limits. It needs knowledge of several enterprises, adds labour and management time, may concentrate disease risk (for example, poultry manure in fish ponds can carry pathogens if poorly managed), and can be constrained by water, land tenure and market access for diverse products. The best combination depends on the farm's resources, the household's skills and the market. Integration is also easier where enterprises are close together, as in homestead systems, because transport of manure, fodder and water is a real cost on small farms without vehicles.
Rice-fish culture raises fish in flooded rice fields; the fish eat weeds and insects, stir the soil and add nutrients, and the farmer gains protein and income, though rice fields must be modified with trenches or refuges and pesticide use must be limited. Rice-duck systems use ducks for weed and pest control. In southern China, the dike-pond system of the Pearl River Delta combined fish ponds with mulberry (for silkworms), sugarcane or vegetables on the dikes, with pond mud returned to the dikes and silkworm waste fed to fish. In Vietnam, the VAC system combines garden, pond and livestock pen around the homestead.
In India, farming system models developed by ICAR institutes and state universities combine field crops with dairy, poultry, goats, fish, horticulture, mushrooms and biogas in proportions suited to each zone, typically for farms of one to two hectares. Reported benefits include higher net income, year-round employment and more recycling than cropping alone, but results vary widely with management and markets, and many published figures come from research farms.
Analysing an integrated farm starts with flows. Manure output can be estimated from animal numbers and standard excretion values, which vary with species, diet and housing; nitrogen in manure is lost as ammonia during storage and spreading, so the nitrogen available to crops is lower than the amount excreted. Feed self-sufficiency is the share of animal feed produced on the farm. Fish production in manured ponds depends on stocking density, species combination (as in composite fish culture with surface, column and bottom feeders) and water quality.
A simple partial budget compares the extra returns of adding an enterprise with its extra costs, including labour. Integrated systems should be judged on whole-farm net income, risk and resource use, not on the output of each component alone, since a fish pond that loses money on its own may still raise farm income by recycling wastes and saving fertiliser. Biogas plants add another loop: animal dung digested anaerobically yields gas for cooking and a slurry that remains a useful fertiliser, though nitrogen can be lost from slurry stored in the open.
Rainfed agriculture relies on rainfall rather than irrigation. FAO estimates that it covers roughly four-fifths of the world's cropland and produces well over half of its food; in India, roughly half of the net sown area is rainfed, supporting most pulses, coarse cereals and oilseeds and much of the livestock population. Rainfed areas also hold many of the world's poor, so their productivity and resilience are central to food security.
Aridity is often classified by the aridity index, the ratio of annual precipitation to potential evapotranspiration (P/PET). Under the UNEP classification, values below 0.05 are hyper-arid, 0.05 to 0.20 arid, 0.20 to 0.50 semi-arid and 0.50 to 0.65 dry sub-humid; drylands are the zones below 0.65. A site with 400 mm rainfall and 1,600 mm PET has an index of 0.25, semi-arid. Indian agronomy texts also distinguish dry farming (areas below about 750 mm rainfall), dryland farming (about 750 to 1,150 mm, with moisture stress during the season) and rainfed farming (above about 1,150 mm), though usage varies.
Rainfed systems face erratic rainfall, dry spells during the season, high evaporative demand, low soil organic matter and fertility, erosion and, often, poor access to inputs, credit and markets. Strategies combine several measures. In-situ moisture conservation (contour cultivation, ridges and furrows, compartmental bunds, mulching and conservation agriculture) keeps rain where it falls. Rainwater harvesting in farm ponds and check dams stores runoff for supplemental irrigation at critical stages. Drought-tolerant crops and varieties, such as millets, sorghum, pigeonpea and chickpea, and adjusting sowing dates to the onset of monsoon reduce risk. Contingency crop plans prepare alternative crops and varieties for delayed monsoons, and intercropping spreads risk.
Watershed development integrates these measures across a landscape, combining soil and water conservation, afforestation, water harvesting and livelihoods; the land degradation and water governance subjects of this track cover it in detail. Research institutions such as ICRISAT and India's Central Research Institute for Dryland Agriculture (CRIDA) at Hyderabad specialise in these systems.
The potential runoff from a catchment is estimated as area x rainfall x runoff coefficient. A 2-hectare catchment (20,000 square metres) receiving 600 mm (0.6 m) of seasonal rainfall with a runoff coefficient of 0.25 yields about 20,000 x 0.6 x 0.25 = 3,000 cubic metres. Pond design must allow for seepage and evaporation losses, which can be large in hot, sandy areas, and the value of stored water depends on using it at the most sensitive crop stage. Supplemental irrigation of even one or two applications at flowering or grain filling often gives large yield gains in rainfed crops.
Runoff coefficients vary with rainfall intensity, slope, soil texture and cover: bare, crusted or rocky catchments shed much more water than vegetated, sandy ones, and coefficients for a single storm differ from seasonal averages. Planners therefore use local measurements or published tables for similar catchments, and check designs against rainfall records for dry years, when stored water matters most. Small farm ponds are often lined or sealed where seepage losses would otherwise make them uneconomic.
Sustainable intensification (SI) means raising agricultural output from the same area of land while reducing negative environmental impacts and increasing contributions to natural capital and ecosystem services. The idea was developed in the 1990s, notably by Jules Pretty, and became prominent after the Royal Society's 2009 report Reaping the Benefits and the 2010 analysis by Godfray and colleagues in Science on feeding nine billion people. Its logic is that land conversion for agriculture is a major driver of biodiversity loss and emissions, so meeting future food demand should rely on higher yields on existing farmland rather than expanding it.
SI is defined by goals, not by technologies: it is neutral about whether higher yields come from improved varieties, fertiliser and precision agriculture or from agroecological redesign, provided environmental impacts fall. Pretty and colleagues (2018) used the efficiency-substitution-redesign framing (see agroecology-principles) and estimated that roughly 163 million farms, about 29 percent of farms worldwide, had adopted some form of SI redesign on about 9 percent of agricultural land, figures based on compiled project data with acknowledged uncertainty.
Critics raise several points. SI is said to have a productivist bias, putting higher output first and environment second, and to neglect distribution: hunger is driven by poverty and access as much as by total production. Others argue that the concept is vague enough to be used to justify business as usual, including heavy input use, under a sustainability label. The Jevons paradox or rebound effect warns that higher productivity can make farming more profitable and encourage expansion into new land unless land-use policy prevents it, so intensification does not automatically spare land. Some critics prefer 'ecological intensification', relying on ecosystem services, or agroecological transformation.
Supporters reply that the concept deliberately includes both technological and ecological options, that yield gaps on many smallholder farms are large and closing them can reduce poverty, and that safeguards for equity and land use can be built in. The useful approach for students is to treat SI as a set of goals and ask, for any proposal, what happens to yield, input use, emissions, biodiversity, land use and who benefits.
A yield gap is the difference between potential (or water-limited potential) yield and actual farm yield. Potential yield is set by climate and variety with no water, nutrient or pest limits; water-limited potential applies to rainfed crops. Large yield gaps in sub-Saharan Africa and parts of South Asia suggest room for intensification, while gaps in intensive systems are small, so further gains there are hard and costly. Projects such as the Global Yield Gap Atlas estimate gaps using crop models and local data. Closing yield gaps sustainably usually requires better nutrient management, improved varieties, timely operations, water management and access to markets and credit.
Yield-gap estimates are uncertain because potential yields come from models and actual yields from statistics or surveys of varying quality. Farmers may also rationally accept yields below the economic optimum because of risk, cash constraints or labour, so the 'exploitable' gap is usually taken as a fraction of the full gap, commonly put at about 70 to 80 percent of potential in intensive systems, beyond which extra inputs rarely pay.
Nutrient use efficiency can be expressed in several ways, and examiners expect students to keep them apart. Partial factor productivity (PFP) is crop yield per unit of nutrient applied: 4,500 kg grain from 120 kg N gives a PFP of 37.5 kg grain per kg N. Agronomic efficiency (AE) is the yield increase per unit of nutrient applied, compared with an unfertilised plot: (4,500 - 2,700) / 120 = 15 kg grain per kg N. Apparent recovery efficiency (RE) is the increase in nutrient uptake per unit applied: if uptake is 95 kg N with fertiliser and 50 kg N without, RE = (95 - 50) / 120 = 37.5 percent. Physiological efficiency is the yield increase per unit of extra nutrient taken up. The nitrogen use efficiency used in national balances (output / input) is a different measure again (see agroecology-principles).
Low efficiency means nutrients are lost to the environment or stored in soil. Improving it depends on the right source, rate, time and place of application (the '4R' approach of nutrient stewardship, covered in its own subject), on healthy crops that can use the nutrients and on balanced fertilisation.
Water productivity is crop output per unit of water used, often expressed as kilograms of grain per cubic metre of evapotranspiration or of water applied. A crop yielding 4,500 kg/ha that uses 450 mm of water (4,500 cubic metres per hectare, since 1 mm over 1 hectare is 10 cubic metres) has a water productivity of 1.0 kg per cubic metre. Field water use efficiency and irrigation efficiency are related terms with different denominators, so units and definitions must be stated. Labour productivity (output per person-day) and land productivity (output per hectare) often move in different directions, which matters when comparing labour-intensive integrated systems with mechanised ones.
Efficiency ratios can mislead when output falls. Using less water or fertiliser per kilogram is good, but a system that halves inputs and yields equally has the same efficiency and produces less food, possibly requiring more land elsewhere. Analysts therefore report absolute output and resource use alongside ratios, and consider whether resource savings are real at the scale of the watershed or aquifer, since water 'saved' on one field may simply have recharged groundwater used by a neighbour.
Farm economic analysis uses enterprise budgets: gross returns (output x price), variable costs (seed, fertiliser, pesticides, hired labour, fuel), gross margin (returns minus variable costs), fixed costs and net returns. The benefit-cost ratio is gross returns divided by total costs: returns of 120,000 against costs of 80,000 give a ratio of 1.5. Indian cost of cultivation studies use cost concepts such as A1, A2, B1, B2, C1, C2 and C3, which add imputed costs of family labour, owned land and capital step by step (treated in the agricultural economics subjects).
Total factor productivity (TFP) compares an index of all outputs with an index of all inputs; its growth measures technical progress and efficiency rather than simply more inputs. Input-output analysis at farm level should also count environmental costs where possible, such as nutrient losses or groundwater depletion, which market prices do not reflect; this is part of what sustainability assessment, in the next chapter, tries to do.
Sustainability has environmental, economic and social dimensions, and many frameworks add governance. Indicators should be relevant to the question, measurable at reasonable cost, sensitive to change, understandable to users and, ideally, linked to thresholds or targets. Common farm-level indicators include soil organic carbon, nutrient balances, erosion, pesticide use and toxicity, water use, energy and greenhouse gas emissions, biodiversity (crop diversity, habitat), net farm income, income stability, debt, labour and working conditions, food security and access to services.
Indicators measured in different units must be normalised before they can be combined. A common method is min-max normalisation: (value - minimum) / (maximum - minimum), giving a score between 0 and 1; a value of 6 on a scale from 2 to 10 scores (6 - 2) / (10 - 2) = 0.5. For indicators where lower is better, such as pesticide use, the formula is reversed. Normalised indicators can be weighted and aggregated into a composite index, but the choice of weights is a value judgement, and aggregation can hide trade-offs, so good practice reports the individual indicators as well, often in radar or spider diagrams.
Several frameworks are used at farm level. FAO's SAFA (Sustainability Assessment of Food and Agriculture systems) organises themes under good governance, environmental integrity, economic resilience and social well-being. RISE (Response-Inducing Sustainability Evaluation), developed in Switzerland, and IDEA, from France, score farms on sets of indicators. MESMIS and FAO's TAPE, which links scores to the ten elements of agroecology, are described in agroecology-principles. FAO's SHARP tool assesses climate resilience of smallholders.
At national level, Sustainable Development Goal indicator 2.4.1, for which FAO is the custodian agency, measures the proportion of agricultural area under productive and sustainable agriculture. It uses a farm-survey approach with eleven sub-indicators covering the economic dimension (such as farm output value per hectare, net farm income and risk mitigation), the environmental dimension (such as soil health, water use, fertiliser and pesticide risks and biodiversity) and the social dimension (such as decent employment, food security and secure tenure). Each sub-indicator is classified on a traffic-light scale as desirable, acceptable or unsustainable using thresholds, and countries report the share of agricultural area meeting the criteria, collected through agricultural surveys.
Assessments go wrong when indicators are chosen because they are easy to measure rather than important, when farms are compared without considering their context, when data are self-reported without checks, and when composite scores are treated as precise. Sustainability is also a direction rather than a fixed state, so repeated assessments over time are more informative than a single score. Participatory assessment, in which farmers help choose and interpret indicators, improves relevance and use of results.
A further pitfall is scale. A farm can score well while shifting impacts elsewhere: buying feed grown on cleared forest, exporting manure surpluses to neighbours, or using groundwater faster than recharge across the aquifer. Frameworks that include upstream inputs, landscape effects or life cycle impacts reduce this risk, at the cost of more data. Analysts should state the boundary of each assessment and avoid treating farm-level scores as measures of sustainability for the whole food system.
Energy budgeting accounts for the energy used to produce crops and livestock and the energy contained in the products. Inputs are divided into direct energy (diesel, electricity, human and animal labour) and indirect energy (energy embodied in fertilisers, pesticides, seed, machinery and irrigation infrastructure). Each physical input is multiplied by an energy coefficient. Commonly used coefficients include about 56 MJ per litre of diesel, about 60 MJ per kg of nitrogen fertiliser (published values range widely, roughly 40 to 80, depending on manufacturing efficiency), about 2 MJ per hour of adult human labour in Indian studies, and about 14 to 15 MJ per kg for cereal grain; coefficients vary between sources and must be stated.
Outputs are the energy content of the harvested products and, if used, by-products such as straw. Studies differ in whether they count straw, labour and machinery manufacture, so results are comparable only within a study or where boundaries match.
Energy use efficiency (energy ratio) = energy output / energy input. Energy productivity = yield (kg) / energy input (MJ). Net energy = energy output - energy input. Specific energy = energy input / yield (MJ per kg). As an example, suppose a wheat hectare uses 60 L of diesel (60 x 56 = 3,360 MJ), 120 kg N (120 x 60 = 7,200 MJ), 300 hours of labour (300 x 1.96 = 588 MJ) and 100 kg of seed (100 x 14.7 = 1,470 MJ), a total of 12,618 MJ, and yields 4,500 kg grain (4,500 x 14.7 = 66,150 MJ) and 6,000 kg straw (6,000 x 12.5 = 75,000 MJ). Counting grain and straw, output is 141,150 MJ, energy use efficiency is about 11.2, energy productivity is 4,500 / 12,618 = 0.36 kg per MJ, net energy is 128,532 MJ and specific energy is about 2.8 MJ per kg of grain. Counting grain only would give an efficiency of about 5.2, which shows why boundaries must be stated.
In most intensive cereal systems, nitrogen fertiliser and diesel are the largest energy inputs, followed by irrigation pumping where groundwater is deep. Energy budgets therefore identify where savings matter most: legumes in rotation, efficient nitrogen management, reduced tillage and efficient pumps.
Energy budgets are useful for comparing systems, identifying energy hotspots and linking to greenhouse gas emissions, since fossil energy use is closely related to carbon dioxide emissions. They are limited because energy is not the only scarce resource: a system with a high energy ratio may use much labour or land, and food quality and nutrition are not captured by energy content. Life cycle assessment, covered in its own subject, extends the approach to many impact categories with standardised methods.
Energy analysis is also sensitive to the treatment of human labour. Some studies count only the metabolic energy of a worker's hours, others the full food energy consumed by the household, and others exclude labour altogether; the choice can change the ranking of labour-intensive and mechanised systems. Comparisons between traditional and modern systems, which are popular in exams, should therefore be read with the coefficients and boundaries in mind rather than as absolute statements about which system is more efficient.
Most of the world's farms are small. Lowder and colleagues (2016) estimated more than 570 million farms worldwide, of which roughly 475 million were smaller than 2 hectares; these small farms operate a modest share of farmland but a much larger share of farms. Ricciardi and colleagues (2018) estimated that farms under 2 hectares produce roughly a third of the world's food supply on about a quarter of agricultural land, with considerable uncertainty. Smallholders are central to food security, rural employment and biodiversity in many countries, and many are also poor, food-insecure and exposed to climate and market shocks. Farm size is also falling in many low-income countries as land is divided between heirs, while it is rising in high-income and some middle-income countries; these opposite trends shape which technologies and policies are relevant.
Smallholder farming systems are usually diverse: several crops, some livestock, trees and off-farm income from wage labour, trade or remittances. Diversification is partly a response to risk, partly to labour availability and partly to the household's need for food, cash and savings.
The sustainable livelihoods framework, widely used since the late 1990s (notably by the UK Department for International Development in 1999), analyses how households use assets to pursue livelihood strategies within a vulnerability context and under the influence of policies and institutions. It identifies five types of capital: human capital (skills, health, labour), natural capital (land, water, biodiversity), financial capital (savings, credit, income), physical capital (tools, infrastructure, housing) and social capital (networks, groups, trust). The vulnerability context includes shocks (droughts, floods, price crashes, illness), trends and seasonality. Transforming structures and processes, such as government policy, markets and customary institutions, shape access to assets. Livelihood outcomes include income, food security, well-being, reduced vulnerability and more sustainable use of natural resources.
The framework helps identify which assets constrain a household and which interventions might help, such as credit, skills training or secure land rights, and it reminds analysts that farm technology is only one part of livelihoods.
Resilience of farming households and systems is often described as three capacities: absorptive capacity (coping with a shock using existing resources), adaptive capacity (adjusting practices to changing conditions) and transformative capacity (changing the system fundamentally when the old one is no longer viable). Measures that build resilience include diversification of crops, livestock and income, savings and group credit, insurance, social protection, access to information such as weather forecasts, and soil and water conservation.
Income diversification can be measured with the same indices used for biodiversity. A Simpson diversity index of income shares, 1 - sum of squared shares, of 0.62 for shares of 0.5, 0.3 and 0.2 shows a fairly diversified household; a household relying on one source scores 0. Crop insurance schemes, such as India's Pradhan Mantri Fasal Bima Yojana launched in 2016, transfer some risk but face problems of coverage, claims and premiums. Diversification has limits too: it can lower returns compared with specialisation, so the right balance depends on risk, markets and the household's goals.
Natural farming is a family of low-input approaches that avoid synthetic fertilisers and pesticides and rely on on-farm biological inputs, mulching and diversity. Masanobu Fukuoka in Japan described 'do-nothing' farming without tillage, fertiliser or weeding in The One-Straw Revolution (first published in Japanese in 1975). Korean natural farming, associated with Cho Han-kyu, uses indigenous micro-organism cultures. In India, Subhash Palekar promoted what became known as zero budget natural farming (ZBNF), built on four components: jeevamrutha, a fermented mixture of cow dung, cow urine, jaggery, pulse flour and soil applied as a microbial inoculant; beejamrutha, a seed treatment from similar ingredients; acchadana, mulching; and whapasa, a soil condition with balanced air and moisture.
Andhra Pradesh launched a state-wide community-managed natural farming programme in 2016. Nationally, India supported natural farming under the Bharatiya Prakritik Krishi Paddhati component of its organic farming scheme from 2020-21, and in 2024 approved a standalone National Mission on Natural Farming. Natural farming in these programmes is usually not certified organic, although it shares the avoidance of synthetic inputs.
Evidence is mixed and still accumulating. Farmer surveys and programme reports often describe lower costs, similar yields and better soil, while some controlled trials at research stations have reported lower yields for some crops, especially nitrogen-demanding cereals, in the first years without synthetic fertiliser. Differences arise from what is compared (natural farming versus conventional practice or versus recommended practice), how long the system has run, how inputs and family labour are costed, and whether the households studied were selected or self-selected. The claim that jeevamrutha can supply crops' full nutrient needs is not supported by its small nutrient content; its effects, if any, are thought to work through soil biology, and they need further testing.
Students should read natural farming studies by asking: what was the comparison, over how many seasons, on which soils and crops, were yields measured or recalled, were labour and inputs fully costed, and was there a control group? Rigorous multi-year, multi-location comparisons are what the policy debate most needs.
The System of Rice Intensification (SRI), developed in Madagascar in the 1980s by Henri de Laulanié, combines transplanting young seedlings (often 8 to 12 days old), single seedlings per hill at wide spacing, intermittent rather than continuous flooding, mechanical weeding and organic matter. Many farmers report higher yields with less water and seed, but some controlled trials found no yield advantage over good conventional practice, and the extent of benefits remains debated; the water-saving element overlaps with alternate wetting and drying.
Low-external-input sustainable agriculture (LEISA) is a broader term for systems that use local resources and minimal purchased inputs. Such systems can reduce costs and environmental harm, but they may need more labour and knowledge, and in nutrient-depleted soils some external nutrients, organic or mineral, are usually needed to raise yields. The Sri Lanka episode of 2021, discussed in agroecology-principles, shows the risks of forcing low-input transitions too fast. Evidence reviews therefore treat low-input systems as options whose performance depends on soil fertility status, crop, labour availability and markets, rather than as universally superior or inferior to conventional systems.
Master the indices and their exact definitions: cropping intensity, land use efficiency, rice equivalent yield, production efficiency, PFP, AE, RE, water productivity, energy ratio, energy productivity and specific energy, and min-max normalisation. Practise each with numbers until you can do it quickly and state the units. Learn the dryland classifications, the sustainable livelihoods framework's five capitals and the three resilience capacities. On a field visit, draw a farm's resource-flow map and seasonal labour calendar, and list which by-products are recycled and which are wasted.
As an extension worker, NGO staff member or farm adviser you design and evaluate whole-farm changes: adding a fish pond or small ruminants to a crop farm, diversifying a rice-wheat rotation, planning water harvesting on a dryland farm, or supporting a natural farming programme. You build enterprise and partial budgets, estimate nutrient and feed flows, and compare options on income, labour, risk and resource use. You also run sustainability assessments with farmer groups and present results in ways farmers can use.
At senior level you shape programmes and policies for farming system development: which farm models to promote in which zones, how to target smallholders, how to measure progress on SDG 2.4.1 or national sustainability goals, and how to judge claims for sustainable intensification or natural farming. You commission evaluations with proper comparison groups, balance productivity with equity and environment, and avoid scaling models that worked only on research farms. You also coordinate across ministries, because farming system outcomes depend on water, credit, markets and land policy as much as on agronomy.
Research questions include quantifying synergies and trade-offs in integrated systems, modelling whole-farm responses to climate and market change, improving indicators and aggregation methods for sustainability assessment, testing natural farming and SRI claims rigorously over many seasons and sites, and analysing the conditions under which intensification spares or consumes land. Good research combines farm surveys, long-term experiments and bio-economic models, reports uncertainty and context, and involves farmers in defining what sustainability means for them. Senior researchers also help national statistical systems collect the farm-level data that indicators such as SDG 2.4.1 require.
Energy coefficients, excretion values, cost concepts and indicator weights vary by source and country; values here illustrate methods and must be replaced by local data. Income and yield results of integrated farming models and natural farming are not quoted as fixed figures because published estimates vary widely and many come from research farms or programme reports. Detailed watershed design, nutrient stewardship, conservation agriculture practices, agroecological frameworks and life cycle assessment are covered in their own subjects. Farm economics is treated only at the level of enterprise budgets; production function analysis and linear programming of farm plans belong to agricultural economics. National schemes after 2025 are not covered.
Original practice questions written to the syllabus and pattern of the exams named in each question's examTags. They are not past papers and do not reproduce any board's questions.
These notes are free to read here, with no account.
What the Laisha app adds:
Free for ten days, then a subscription. We would rather write that here than have you find it later.