Sustainable Warm-Climate Forage Legumes: Versatile Products and Services

Abstract

Forage legumes, besides their use as ruminant feed supplements, contribute to other agricultural, forestry and natural ecosystems’ sustainability around the world. Our objective in this summary is to emphasize that versatility in the face of biotic, abiotic and socio-economic variability is among the most important traits that forage legumes contribute to sustaining human populations in those diverse ecosystems. Forage legumes could contribute even more to agroecosystems if we 1. consider ecosystem services as well as food, feed and fuel production; 2. more fully exploit what we already know about forage legumes’ multiple uses; and 3. focus greater attention and energy exploring and expanding versatility in currently used and novel versatile species. To draw attention to the importance of this versatility to sustainable grasslands, here we review multiple legumes’ roles as forage, bioenergy, pulses (legume seeds for human consumption), pharmaceuticals and cover crops as well as environmental services, in particular soil health, C sequestration and non-industrial organic N. The major points we single out as distinguishing sustainable versatile forage legumes include (1) multiple uses; (2) adaptation to a wide range of edaphoclimatic conditions; (3) flexible economic contributions; and (4) how genomics can harness greater legume versatility. We predict that, because of this versatility, forage legumes will become ever more important as climates change and human pressures on sustainable agro-environments intensify.

1. Introduction

The most widely used sustainable agricultural and conservation legumes around the world are those with multiple applications arising from versatile biotic and abiotic adaptations. This versatility makes them invaluable in ecosystems with increasingly unpredictable natural and human conditions. As a result, legumes, encompassing 800 genera and 20,000 species, are the second most important crop family [1]. Current and future domestication, plant breeding, conservation and physiological research on legumes could produce even greater impacts if they take full advantage of this versatility.

Versatility is the opposite of brittle species that are narrowly adapted and easily disturbed by changing biotic and abiotic conditions. Examples of legume versatility abound. Some species started out historically as forages and are used today for direct human consumption. Alfalfa (Medicago sativa L.) is a prime example of this adaptability. Originally domesticated as a forage for cattle and horses, today it is also used as salad sprouts and nutritional supplements for human consumption, bioenergy and biotechnology [2]. Soybean (Glycine max L.), originally from China, and cowpea (Vigna unguiculata L.) from Africa are the opposite, domesticated originally as pulse (legume grains consumed by humans) and vegetable (leaves) crops for human consumption [1], but today they serve as forages and cover crops as well.

These multiple-use legumes include species that are herbaceous, arboreal and everything in between. Herbaceous pulse and forage species are by far the most common around the world, with species adapted to tropical, subtropical and cold climates. Leguminous trees are less widely included in silvopastoral agroecosystems but can be crucial in drier, warmer climates because their taproots reach sub-soil moisture crucial for dry seasons or droughty year survival [3]. Bushy perennials are the least commonly domesticated, although they are extremely important for domesticated and wild browsers, especially in arid and semi-arid environments with unpredictable shallow-soil moisture where short-rooted herbaceous species are not dependable forage sources. Perennial legumes especially can contribute to sustainable forage and pulse agroecosystems through stable resource use, climatic resilience and dependable productivity [4]. Exploring more arboreal and brushy legumes, which are almost exclusively perennial, for multiple applications will be crucial in fully exploiting legume versatility into the future.

Legume versatility lends itself to wider ranges of biotic stresses, such as pests and pathogens. This is particularly important in crops that are cultivated over wide regions and varied agroecosystems. Cowpeas, for example, have a wide genetic variability that allows plant breeders and farmers to select which are resistant to myriad biotic stresses, including bacterial blight (Xanthomonas campestris), cowpea aphid-borne mosaic virus (CABMV), cowpea aphid (Aphis craccivora), cowpea curculio (Chalcodermus aeneus), root-knot nematodes (Meloidogyne incognita and M. javanica), cowpea weevil (Callosobruchus maculatus), Striga gesnerioides and Alectra vogelii [5].

Grassland legume atmospheric N₂ fixation (often referred to as biologically fixed N₂) potential is a critical grassland legume contribution. Not only do legumes cooperate with microbes to fix N₂, but they indirectly contribute crude proteins to soil microbiota as well as neighboring forbs and grasses. Biological N₂ fixation varies with legume species, legume proportion in the botanical composition, ecological conditions and management, with average values found across different systems ranging from 50 to 150 kg N ha⁻¹ yr⁻¹ [6,7]. Legumes enhance nutrient cycling by producing an N-rich forage, which can return N to the grassland ecosystem either via excreta or litter [7,8,9,10]. For example, in Brazil, mixed pastures of Desmodium ovalifolium and creeping signalgrass (Brachiaria humidicola) resulted in greater litter N concentration, a reduced C/N ratio in the microbial biomass, and greater soil N mineralization compared with the grass monoculture [11].

Table 1. Atmospheric N₂ contribution potential from grassland legumes.

Legume	System	Climate	Duration	N Contributed	Citation
Arachis pintoi	Forest biome	Tropical	8 years	72.5 kg N ha−1 yr−1 in ecosystem	[13]
Aeschynomene americana	Savannah	Subtropical	3 years	9 kg N ha−1 yr−1in legume	[16]
Pisum sativum	Winter annual grass mix	Subtropical	2 years	213 kg N ha−1 yr−1 in legume	[17]
Vigna radiata	Summer annual grass mix	Subtropical	2 years	167 kg N ha−1 yr−1 in legume	[17]
Trifolium spp.	Grasslands	Temperate	4 years	27 kg soil mineral N ha−1 0 to 60 cm	[14]
Calopogonium muconoides	Brachiaria brizantha + C. muconoides mixed pasture	Tropical	2 years	73 to 98 kg N ha−1 yr−1 from biological nitrogen fixation	[7]

shows a few of the many examples of just how important this N contribution can be to grassland ecosystems. Not only does this N addition benefit grassland ecosystems directly by accumulating crude protein in legume herbage, it also contributes indirectly to soil C accumulation by generally increasing herbage below- and above-ground biomass [12,13]. The advantage of using legumes to supply N to grasslands is that much less nitrate leaching and ammonium volatilization occurs vis-à-vis industrial N fertilizer application [14,15].

Abiotic versatility in legumes is particularly important in dry climates with widely varying environmental conditions. The most widely adapted legumes will contribute something every season, regardless of abiotic conditions. For example, legumes that produce something regardless of stress level can serve as survival crops under conditions of unpredictable rainfall within seasons, among growing seasons, across a range of latitudes, altitudes or longitudes, or long-term changes resulting from climate changes [18,19]. Wide genotype variability means a wider adaptation to the environment through phenotypic expression, especially of traits that adapt to stresses (

Table 2. Legume traits that lend themselves to adaptability.

Trait	Range	Biotic Factors	Abiotic Factors	Examples
Architecture	Dense to open	Plant competition	Wind	Rangeland
Canopy	Small to large	Plant competition	Light interception
Habit	Decumbent to upright	Grazing	Soil moisture (added this based on the general trend of shorter stature plants in arid environments)	Rangeland
Root depth	Shallow to deep	Herb vs. tree	Soil texture Water table Rainfall distribution	Rangeland
Height	Decumbent to upright	Seedling populations	Soil texture Water table Rainfall distribution	Pulses
Climbing	Viney to upright	Plant competition Grazing (exposed meristem)	Light interception	Cowpeas
Flowering	Early to late	Seedling populations	Rainfall distribution	Soybean days to flowering
Seed set	In/determinant	Herbivory/harvest schedule	Soil moisture	Cowpea
Seed quantity	Low to large	Herbivory/haying	Climate/soil	Annuals vs. perennials
Seed size	Small to large	Ecological succession	Climate	Annuals vs. perennials
Rainfall requirements	200 to 2000 mm	Annual vs. perennial	Climate	Days to flowering
Heat tolerance	Up to 40 °C	Leaf abscission	Altitude/latitude	Forage legumes
Cold tolerance	Freeze vs. tropical	Life cycle	Altitude/latitude	Forage legumes
Propagation	Seed vs. vegetative	Herbivory	Rainfall	Perennial peanut

). For example, the faba bean (Vicia faba L.) has sufficient genotypic variability that its phenotypic expression can handle a wide range of abiotic stresses [20]. This genotypic variability allows these legumes to respond positively as abiotic conditions improve. Thus, increased yields occur during high precipitation years or when greater agricultural inputs such as soil amendments and irrigation are available. When conditions are not ideal, however, they still contribute without total loss. As climates change, versatility can be important to permit adaptation to various factors, such as increased atmospheric CO₂ levels in the field bean (Phaseolus vulgaris L.) [21].

Versatility caters to socio-economic as well as biotic and abiotic variability and grassland sustainability. Markets change and farmer resources vary; legumes that meet these varying demands will be the most widely applicable across a gamut of socio-economic conditions. While commercial farmers may be interested in once-a-season mechanical cowpea pulse harvest, their subsistence neighbors may want to utilize the same legumes for their family’s daily hand-picked vegetable consumption over an entire season [22]. Likewise, some farmers or consumers may be interested in cowpea or groundnut (peanut; Arachis hypogea L.) as pulses while others may prioritize dual pulse and forage characteristics [23]. Market-driven multiple uses for legumes include forage, fuel, lumber, soil improvement, erosion control, shade, pulses, vegetables and pharmaceutical products, among myriad others.

Legume diversity under diverse socio-economic conditions indicates that positing too much dependence on a few pulse or forage legume species for humanity’s needs is risky. Novel [24] or underutilized [25] legume species can be further explored for their versatility to environmental stresses and human-related uses, particularly as dietary proteins and bioactive compounds. Their applicability to other uses, such as enhancing soil health or ecosystem services (ESs), could also be explored in agronomic as well as environmental fields [26,27].

Likewise, exploring greater genetic breadth in existing domesticated legume species, namely cultivars or ecotypes already widely used, can broaden as well as safeguard a legume species’ applicability. For example, some determinate cowpea cultivars are suited for pulse production, while fresh leaves and pods in indeterminate types are widely utilized for human consumption and ruminant feed [22]. Harnessing wide genetic variability in wild populations prior to domestication is important in maximizing legume versatility, since domestication has historically involved only a small portion of the genetic diversity in wild germplasm. This is the case, for example, with Medicago spp. [28] and Arachis spp. [29,30]. Wider use of genomic tools could be a key to improving adaptability in legumes already in wide use. Unless we catalogue entire genomes, we will not know what is available.

Each of these forage legume characteristics contribute to sustainable agroecosystems; however, we propose that considering such contributions as flexible (the opposite of brittle), responsive (as opposed to focused on one facet) and multiple needs to be recognized if we are to fully utilize forage legume potential. Our objective in this forage legume apologetics is to present evidence that supports greater versatile exploration and use of these grassland plant species in maintaining ecologically and agriculturally sustainable pastures, rangeland and native prairies. We do this by reviewing the literature as well as proposing ideas and research topics that expand forage legume versatility in sustainable grassland ecosystems.

2. Pulses for Food and Pharmaceuticals

Versatile legumes can be used for multiple purposes, depending on changing farmer priorities. Direct seed and leaf consumption provide important protein and mineral sources for most populations around the world. Pulses are the edible seeds of grain legumes consumed by humans as food in the form of mature dry seed, and in some cases as unripe pods, immature seed or young succulent leaves as an important source of dietary protein, fiber, minerals, vitamins and several essential health factors. The term “pulse” derives from the Latin words puls or pultis, meaning “thick soup” or “thick gruel, porridge, mush” [31]. Most pulses are used as thick soup with different names in many countries, for example, ‘dahl’ in India, which is the largest producer and consumer of pulses in the world. Archaeologists have discovered traces of pulse production by the Indus Valley Civilization dating back to 3300 BCE. Evidence of cultivation of lentils has also been found in the Egyptian pyramids and dry pea seeds have been discovered in a village in Switzerland dating back to the Stone Age. Archaeological evidence suggests that peas were grown in the eastern Mediterranean and Mesopotamia regions at least 5000 years ago [31].

The evolutionary history of major protein-rich pulses evolved in tandem with starch-rich cereals and root crops in different geographical regions, and they together ensured a nutritionally balanced diet for humans irrespective of their place of settlement [31]. For example, (i). wheat, barley, pea, chickpea and lentil evolved in West Asia; (ii). rice, pigeon pea, mung bean in India; (iii). rice and soybean in China; (iv). sorghum and cowpea in Africa; (v). maize and field beans in Mexico; and (vi). potato, sweet potato, cassava and peanut in South America.

2.1. Primary Pulses

The FAO [32] recognizes 11 primary pulse groups comprising over 30 crops/variety types, but the most widely grown and consumed are common field beans, garden pea (Pisum sativum), chickpea (Cicer arietinum), cowpea, pigeon pea (Cajanus cajan), lentil (Lens culinaris), mung bean (Vigna radiata), bambaranut (Vigna subterranean), faba bean and adjuki bean (Vigna ungularis). Soybean and groundnut (Arachis hypogea) are also widely consumed pulses, but these have been traditionally classified as oilseeds because of their high oil contents. On a worldwide basis, pulses are produced and consumed in over 170 countries. The total global pulse production in 2017 was 96 million Mg, of which beans accounted for 31 million Mg, dry peas 16 million Mg, chickpea 15 million Mg, cowpea 8 million Mg and lentils 7 million Mg [25].

2.2. Nutritional Properties

Pulses are among the most important sources of protein, starch, dietary fiber and many health-promoting factors. Depending on the crop and varieties, these contain, on average, 18.5–40% protein, 32–52% starch, 2–40% fat and 6–16% dietary fiber, as well as vitamins such as niacin, riboflavin and thiamine and minerals such as calcium, iron, potassium, manganese and zinc [33,34,35]. Thus, pulses have great importance in human diets around the world due to their contribution towards meeting a major part of the daily requirements of proteins, carbohydrates, lipids and fibers, as well as essential minerals and vitamins. In addition, the valuable contribution of bioactive compounds including fibers, isoflavonoids, phenolic acids and polyphenols, folate and vitamin B6, alkaloids and protein fractions aid in the prevention of various diseases such as cardiovascular diseases, cancer, diabetes and Parkinson’s disease. Pulse starch, proteins, galactomannans and xyloglucans can also be used as a binder, excipient emulsifiers, thickeners and dispersants in the formulation of various drugs in the pharmaceutical industry.

2.3. Food Uses

There is a great deal of diversity in food uses of individual pulses based on their production potential in different geographical areas. For example, dry peas are popular in Europe; chickpeas, lentils and faba beans in West Asia; chickpeas, pigeon peas, mung beans and beans in South Asia; soybeans and adjuki beans in East Asia; cowpeas and Bambara nuts in Africa; and beans and groundnuts in the Americas [32]. The types of local dishes and specialty products vary from country to country as well as from region to region. The most popular pulses with many types of food preparation are the chickpea in Asia, cowpea in Africa and beans in the Americas.

Pulses are easy to prepare in a variety of foods as versatile components of different types of entrees, salads, snacks, breads, dips and desserts [36]. In the Middle East, India and the Mediterranean, the use of pulses is common in salad, soup, pilaf, köfte, or mixed with meat. Chickpeas are popular as snacks after roasting. Falafel is another traditional Middle Eastern food that is made by using ground chickpeas and/or faba beans in deep-fried balls or patties. Hummus is a popular chickpea- and tahini-based spread that originated from Middle Eastern cuisine. Similarly, cowpea is used in many types of food preparations in Nigeria and other parts of sub-Saharan Africa. The most popular foods made from cowpea are akara (deep fried balls), moin-moin (bean cake) and boiled beans with rice. In India, most of the pulses are split and prepared as ‘dahl’, but cowpea and chickpea are primarily eaten whole as spiced boiled beans (chole) with rice or chapati (wheat tortillas).

Pulses have been used for protein enrichment in pasta and bread, and they also are suitable ingredients in gluten-free foods. Wet and dry fractionation methods, as well as bioprocessing such as germination and fermentation, provide useful tools for novel functional pulse ingredients. Pulse use will likely increase in the future. Especially in combination with cereal, they may find new applications, meeting the sensory and nutritional needs of consumers worldwide [36]. There is growing consumer interest in high-protein and high-fiber pulses in Europe and the USA, which has opened new opportunities for food innovations using pulse ingredients. Pulses are already being recommended in many dietary guidelines in developed countries, such as ‘Health Canada Eating Well’, the ‘MyPlate’ system of the USDA, the ‘Eatwell Plate’ of the UK, and the ‘Nutrition Australia’s Healthy Living Pyramid’. In addition to the increased use of pulses in daily diets, including for protein enrichment of pasta and bread, the fast-food industries in the USA are already promoting pulse-based meat replacers.

2.4. Milk Substitutes

Historically, humans have consumed pulse extracts instead of or as a replacement for ruminant milk. Wilkinson (2011) [37] argues that plant protein efficiency is far greater in the former, consumed directly by humans, compared to the latter, consumed by ruminants to produce milk for human consumption. In situations where pulses can be grown directly for human consumption, efficiency favors them over pulses for ruminant consumption. However, there are many grasslands that are suitable for forage legume production and not for pulse cultivation, making the former system far more sustainable. If soils and climate allow a choice between forage legumes for milk production and pulses for milk replacement, the latter is likely more directly efficient and sustainable for human consumption. Today, that trend is increasing, with various pulse species proposed [38]. Many of these are considered healthier for human consumption than widely used soy extract. As a human food movement, this may prove to be more sustainable than dairy production, thereby eventually reducing grassland use for ruminant milk production for human consumption.

2.5. Leafy Vegetables

Although many forage legumes can also produce pulses consumed as food made of dry seeds, some of them are also consumed as leafy vegetables or fed to livestock as forage [39]. This multiple use enhances forage system sustainability. Young leaves and shoots of chickpeas and peas are widely cooked and consumed as vegetable greens in India and Nepal, and young cowpea leaves are a primary source of vegetable throughout eastern and southern Africa [40,41,42,43]. The young leaves of these pulses have a high nutritional value, including 25 to 30% protein, and are very rich in vitamins, minerals and various health-beneficial phytochemicals, including beta-carotene, vitamin C, folate and fiber. The leaves are normally consumed as fresh relish and boiled vegetables but are also dried and preserved for use in the dry season when foods and vegetables are in short supply [44]. In parts of southern Africa, the cowpea leaves are cooked, pounded and rolled into small balls and dried. This product is called ’Morog’ and is widely marketed and consumed in the region [42]. In this form, leaf products can be kept in the house for a long time before being consumed by the family or sold in the local markets. Cowpea leaves are also dried and stored. Most of the households in eastern and southern Africa have a few late-maturing indeterminate cowpea plants in their backyard as a convenient and fresh source of leaves for stews and soups.

2.6. Pharmaceuticals

In addition to being an important source of proteins, carbohydrates, resistant starch, fibers, minerals and vitamins, pulses contain bioactive compounds which have a positive impact on human health, and therefore, are used as health foods and in drug formulation [35,45,46]. In addition, starch, protein and fibers in pulses can be exploited as binders, excipients, thickeners and dispersants in the formulation of various products in the pharmaceutical industry. Hydrophilic phytochemicals, such as ascorbic acid, phenolic acids and polyphenols, have been associated with enhanced immune system functionality and reduced cancer risk, whereas lipophilic phytonutrients, such as carotenoids and tocopherols, prevent the risk of cardiovascular diseases and some eye pathologies. A positive association exists between the consumption of pulses such as beans with phenolic compounds, especially flavonoids, and the reduction of cardiovascular diseases, glycemic control and diabetes, and the reduction of the risk of pancreatic, kidney and breast cancer [35,45,47,48,49].

Legumes contain phytoestrogens with broad biological activities that are now being applied to humans as treatments for menopause and osteoporosis [50]. Isoflavonoid extracts (genistein and daidzein), particularly prevalent in the Fabaceae subfamily of the Leguminosae, are used as alternative compounds for hormone replacement therapy (HRT) for menopausal disorders. As a result of their superior nutrient content and other health-promoting properties, pulses can play a role in several special diets, such as ‘gluten-free’, ‘diabetic’, ‘vegetarian’ and ‘weight management’ diets [51]. For example, cowpea consumption has many health benefits [52], including anti-diabetic, anti-cancer, anti-hyperlipidemic, anti-inflammatory and anti-hypertensive properties.

3. Forages

Legumes have long been sought as potential forages to integrate into livestock systems. The primary reasons include their ability to associate with soil bacteria to fix atmospheric dinitrogen and thus add N to grasslands, and because of their greater nutritive value. Certain regions of the world are centers of origin of important forage legumes. In the Brazilian northeast, for example, native rangeland vegetation (Caatinga) has a great diversity of legumes, with many presenting forage potential, including Desmanthus, Bauhinia, Clitoria, Stylosanthes, Macroptilium and Arachis [52].

Legume integration into livestock systems remains a challenge, although successful examples are found around the globe [53]. Socio-economic factors important for success include the development of partnerships among researchers, extension workers and farmers; meeting the needs of farmers; understanding the social-economic aspects of local communities and involving community members in the development process; ensuring seed supply and the participation of early adopter farmers who champion the process.

Successful integration of legumes into grazing systems involves the selection of appropriate legume species and livestock management practices. Low forage legume persistence occurs when highly palatable legumes with limited grazing tolerance are used under continuous grazing conditions. Reasons for low grazing tolerance include the exposure of meristems to grazing, reducing the ability of legumes to regrow. A classic example is Stylosanthes guianensis, which typically has meristems more exposed to decapitation under grazing or harvest regimes compared with Arachis pintoi [54].

Tree legumes are a potential alternative to develop sustainable livestock systems [6]. They typically persist longer in the system compared with herbaceous legumes. They can add N via biologically fixed nitrogen (BFN), improve forage protein, provide shade for livestock and deliver other ES [3,55,56].

3.1. Monocultures

Legume monocultures, compared to mixtures with grasses, are more easily maintained with intensive cultivation systems such as hay. A typical example is alfalfa (Medicago sativa L.), which is grown on 32 million hectares around the globe, with 70% of that area occurring in the USA [57]. Such extensive areas of legume monocultures have risks, however, including widespread pests and diseases. The use of legume monocultures in crop rotations, however, might be an option to break the cycle of grass monoculture, or in many cases, continuous cropping of maize, which is the case for many dairies in the USA. Other legumes (e.g., Lespedeza cuneata L.) have minor areas of monoculture compared with alfalfa.

3.2. Mixtures with Grasses

Grass–legume mixtures that are persistent over time are the ultimate goal for many livestock systems. Grasses demand large amounts of N to grow and thrive; legumes provide that. Mechanisms of N transfer from legumes to companion grasses include livestock excreta, litter and below-ground tissue turnover [8]. Bahiagrass (Paspalum notatum Flügge) and rhizoma peanut (Arachis glabrata Benth.) is an example of a successful grass–legume mixture. Because rhizoma peanut is tolerant to grazing, it typically persists over time, although grazing tolerance might vary with cultivar. Garay et al. (2004) [58] assessed two cultivars of rhizoma peanut, ‘Arbrook’ and ‘Florigraze’, under grazing. After 3 years, Florigraze maintained its botanical composition, while Arbrook, an upright cultivar less tolerant to grazing, declined with time. Jaramillo et al. (2018a) [59] established different rhizoma peanut cultivars (Florigraze and Ecoturf), one A. pintoi cultivar (‘Amarillo’) and one A. hypogea, on a previously established Pensacola bahiagrass. After 3 years, Ecoturf was the best legume and reached 30% in the mixture. In a similar trial, but using Tifton-85 bermudagrass, the legumes tended to dominate the mixture after 3–4 years [60]. Another mixture that has been successful in warm-climate regions is palisadegrass [Urochloa brizantha (Hochst. ex. A. Rich.) R. Webster] and A. pintoi.

Avoiding N fertilizer is important in maintaining legumes in mixtures because N favors grass growth. Grazing management of these mixtures is likewise key to maintaining the balance in the botanical composition, with more intensive grazing helping to maintain low-growing legume proportions, such as pintoi peanut [61]. In many cases, the integration of forage legumes reduces the need for N fertilization, resulting in similar forage productivity as N-fertilized all-grass systems [62]. In grass–legume mixtures, companion grasses will compete for soil N, providing a feedback mechanism for legumes to fix more N. Grass–legume mixtures can yield more nitrogen than legume pure stands due to mutual stimulation of nitrogen uptake from BFN and non-symbiotic sources [63].

3.3. Fodder Banks

Legume fodder banks are an option to provide forage with a greater crude protein concentration during a limited browsing time during the day. This management practice might enhance protein nutrition of grazing livestock, especially during dry seasons when dormant low-protein C₄ grasses are the only forage source. There are numerous examples of legume fodder banks in the literature, but most of them refer to arboreal/shrub legumes [64]. Examples of legumes include Leucaena leucocephala (Lam.) de Witt., Gliricidia sepium (Jacq.) Kunth ex Walp. and Cajanus cajan (L.) Millsp. [65]. Fodder banks are also commonly used for greenchop in zero-grazing systems in Africa and Southeast Asia. They are also planted along the edges of rice paddies or as live fences [3].

3.4. Silvopasture

Tree legumes are an emerging option in silvopasture systems (SPSs). Bnefits of using tree legumes as the arboreal component include greater primary productivity, BFN, and provision of more digestible protein for livestock [3]. In many cases, however, secondary compounds, such as condensed tannins, reduce the benefit of protein, since they precipitate in the rumen as well as inhibit animal consumption due to palatability issues [66,67]. In Northeast Brazil, Apolinário et al. (2015) [68] assessed the potential of two tree legumes, Mimosa caesalpiniifolia Benth. and Gliricidia sepium, in an SPS. They observed that Mimosa could increase income by providing marketable wood. In addition, legumes provide shade for livestock and cycle nutrients from deeper soil layers. Livestock performance on these systems was similar during the initial 3 years compared with signalgrass (Urochloa decumbens Stapf.) monoculture [69], but as the trees grew, the competition with the herbaceous vegetation increased, reducing livestock performance in the SPS, especially with Mimosa [70]. Trees might compete for light, water, and nutrients, affecting productivity and quality of the herbaceous component. At the same time, trees can deposit N-rich litter benefiting the vegetation in the understory [71].

4. Bioenergy

Energy is very critical in socio-economic development [72]. In the last 10 to 15 years, there has been a growing interest in bioenergy and biofuels aimed at reducing dependence on fossil fuels [73]. Most current energy systems are heavily dependent on fossil fuels (coal, oil and gas). In 2017, around 80% of the world energy consumption came from these sources, which are considered non-renewable. Of the remaining 20%, about 77% comes from bioenergy (renewable sources), 87% of which comes from wood biomass, 9% from agricultural crops and by-products, and 4% from organic waste from municipal and industrial sources [74].

Renewable energy technologies are ones that consume primary energy resources that are not subject to depletion. Examples include solar, wind, geothermal and biomass energy [75]. The FAO [76] defines bioenergy as “the energy generated from the conversion of solid, liquid and gaseous products derived from biomass (any organic matter, i.e., biological material, available on a renewable basis)”.

Bioenergy from wood is one of the oldest forms and was long considered a primitive energy source in many industrialized countries [77]. However, recurrent energy crises generated growing interest for alternatives to fossil fuels, presenting lignocellulosic materials as promising resources for sustainable energy [78]. Biofuels are fuels produced from biomass, mostly in liquid form, within a time frame sufficiently short to consider that their feedstock (biomass) can be renewed, contrarily to fossil fuels [79]. In this way, legumes represent important forest resources for the energy needs of the population and industry, such as firewood, charcoal, waste and oils. This is because plants of this family are naturally occurring in the tropics, especially in the Americas, Africa and Asia, and provide their own BFN, which can contribute N to companion crops.

Legumes as bioenergy crops contribute several unique functions and ES vis-à-vis other species. These are primarily perennial herbaceous, shrub and arboreal species grown as sole crops or mixtures with grasses to produce biomass for biorefinery and simultaneously building soil organic matter and fertility without the requirement for N-fertilization [80]. Novel non-food leguminous species should be sought to produce bioenergy. Fast growing tree species, for example, Acacia spp., Gliricia sepium (Jacq.) Walp., Leucaena spp. and Prosopis spp. [81,82,83,84], are useful. Leguminous trees have been proposed for biogas and electricity production [85,86]. Biodiesel from legume seeds, for example, Pongamia pinnata (L.) Pierre. (also called Millettia pinnata (L.) Panigrahi), is another alternative in energy production ([83,85,87,88,89,90,91,92].

Several legume species are mentioned in the literature as fast growing and useful for bioenergy production.

Table 3. Examples of potential legumes for bioenergy production (fuelwood, firewood and charcoal) listed.

Species	Habit	Origin
Acaciella angustissima (Mill.) Britton & Rose	Tree	North, Central and South America
Acacia auriculiformis A.Cunn. ex Benth.	Tree	Asia and Oceania
Acacia leptocarpa A.Cunn. ex Benth.	Tree	Oceania
Acacia mangium Wild.		Asia and Oceania
Acacia mearnsii Wild.	Tree	Oceania
Acacia nilótica (L.) Delile	Tree	Africa and Asia
Senegalia polyacantha (Willd) Seigler & Ebinger	Tree	Africa
Senegalia senegal (L.) Willd.	Shrub	Africa and Asia
Aeschynomene histrix Poir.	Shrub	Central and South America
Albizia lebbeck (L.) Benth.	Tree	Africa and Asia
Brachystegia boehmii Taub.	Tree	Africa
Brachystegia spiciformis Benth.	Tree	Africa
Cajanus cajan (L.) Huth.	Shrub	Asia
Calliandra calothyrsus Meissner.	Tree	Central America
Dichrostachys cinerea (L.) Wight & Arn.	Shrub	Africa and Asia
Erythrina excelsa Baker.	Tree	Africa
Gliricidia sepium (Jacq.) Walp.	Shrub	Central America
Inga edulis Mart.	Tree	Central and South America
Leucaena collinsii Britton & Rose	Tree	Central America
Leucaena leucocephala (Lam.) De Wit.	Tree	Central America
Mimosa scabrella Benth.	Tree	South America
Mimosa caesalpiniifolia Benth.	Tree	South America
Pithecellobium dulce (Roxb.) Benth.	Tree	Central and South America
Pongamia pinnata (L.) Pierre.	Shrub	Asia
Prosopis alba Griseb.	Tree	South America
Prosopis juliflora (Sw.) DC.	Tree	Central and South America
Senna siamea (Lam.) H.S. Irwin & Barneby	Shrub	Asia
Senna spectabilis (DC.) H.S. Irwin & Barneby	Tree	South America
Sesbania bispinosa (Jacq.) W.Wight	Shrub	Africa and Asia
Sesbania sesban (L.) Merr.	Shrub	Africa, Asia and Oceania
Tamarindus indica L.	Tree	Africa
Tephrosia candida (Roxb.) DC.	Shrub	Asia
Tephrosia vogelii Hook.f.	Shrub	Africa

Source: Information compiled from [93,94,95,96,97,98].

shows some of the species according to their habits and geographic origins, and which are disseminated in several countries.

5. Cover Crops

Cover crops are defined as plants grown between the last harvest and the next crop planting for the purpose of enhancing ESs within the agroecosystem; however, they may also be grown between rows as living cover in some agroecosystems [76]. Row crops may be planted directly into cover crops in no-tillage or strip-tillage systems or may be terminated with herbicides or equipment to roll and crimp (break the cover crop stem) or till the soil prior to planting the target row crop. Alternatively, some plant species used as cover crops can be harvested for grain, pulses or hay, or grazed by livestock, thereby providing direct income from the cover crop in addition to the row crop. Legumes make ideal cover crops because of their multiple functions within agroecosystems.

Cover crops provide ES similar to those provided by forage crops [99]. Enhanced ESs attributed to cover crops include the following: reduced soil erosion, improved soil quality or health, increased nutrient cycling, broken pest cycles (insects and pathogens), increased small mammal diversity and, specific to the use of legumes, increased N availability in the soil [100,101,102,103,104,105]. The positive effects of cover cropping come with the cost of soil moisture uptake; therefore, cover cropping may not be ideal in all environments, especially where moisture is a limiting factor [106].

Cover crops usually improve soil quality or health, defined as the capacity of the soil to function “to sustain biological productivity, maintain environmental quality, and promote plant and animal health” [107,108,109,110]. Soil health is dynamic and can be measured using methods which assess the physical, chemical and biological status of the soil [111]. For this reason, there is no one measure to define soil health; however, soil organic matter, particularly labile or active pools of carbon, influences physical, chemical and biological components of soil and is considered a key indicator of soil processes [107,109,112]. Increasing soil organic matter improves soil aggregate stability, water infiltration and water holding capacity. The primary driver of increased soil organic matter is additional return of plant biomass to the soil, accomplished through increasing target row crop residue or adding cover crops. Legume cover cropping in a corn (Zea mays L.) and soybean system increased soil carbon storage by 6.6 g m⁻² over 13 years compared to a conventional system with no cover cropping, which had no net change in C storage [113]. In the southeastern USA, a mixture of hairy vetch (Vicia villosa Roth) and rye (Secale cereale L.) produced greater herbage mass than either species alone as a cover crop in cotton (Gossypium hitsutum L.) or sorghum [Sorghum bicolor (L.) Moench)] cropping systems, which is often the outcome of including legumes in grass monocultures [114].

A C/N ratio greater than 25:1 of cover crops may result in immobilization of N and subsequent reduction of target row crop yields [115]. This is where legume cover crops can excel. Nitrogen is provided to the system through legume roots and remaining surface litter decomposition by soil microbes, decreasing the need for inorganic nitrogen input into the system [100,116]. Non-legume or grass cover crops tend to supply greater biomass, and thus greater C; however, legumes fix and supply N, so a mixture of grass and legumes is ideal to both provide biomass and reduce the C/N ratio for N mineralization and availability to the target row crop [115]. The N amount provided by legume cover crops is dependent upon the amount of biomass produced and the efficiency of atmospheric N² fixed biologically, so nitrogen fertilizer equivalences may range from as little as 15 to as much as 200 kg ha⁻¹ yr⁻¹ [117].

6. Soil Health and Fertility

Versatile legumes can promote soil health, mitigate environmental issues and, as green manure and cover crops, provide N to crops that do not have BFN capabilities. Climate change can impact agricultural systems through increased CO₂ levels and greater variability in weather patterns, so legume versatility vis-á-vis soil health and climate change becomes even more important. Secondary impacts of these changes on soils include longer droughts, increased soil salinity, more rapid soil degradation and enhanced pest and disease incidence. Addressing these factors with legumes will necessitate changes in food production practices and input use, which will affect the cost of food production and disproportionately impact smallholder farmers [118]. Use of microbial inoculants designed specifically to bolster legume versatility potentially provides lower-cost alternatives to synthetic inputs. Mutualistic plant relationships with rhizobia, arbuscular mycorrhizal fungi (AMF) and plant growth-promoting rhizobacteria (PGPR) can enhance legume resistance to these current and future environmental stresses [119,120].

Under natural conditions, legumes contribute to complex systems consisting of a variety of relationships and functions. Rhizobial inoculation at seeding has been recommended since the end of the nineteenth century [121], primarily to enhance BFN [122]. Rhizobia not only fix atmospheric N₂ but can also solubilize P and produce enzymes that promote root growth, facilitate mycorrhizal inoculation and protect against plant disease infestation [123]. Plant benefits from rhizobia are enhanced through interactions with AMF and PGPR [124,125]. AMF can grow through fine pores in the soil, extending by over 100 times the root access to soil water and minerals. When rhizobia and AMF are co-inoculated, this can result in greater nodulation and nitrogen fixation [126,127].

Plant metabolites and nutrients from sloughed-off roots and mycorrhizae provide food, while plant roots provide a habit for a diversity of rhizosphere and endophyte bacteria. Approximately 2–5% of these bacteria are referred to as PGPR since they release organic acids, signal chemicals, pheromones or phytohormones that enhance plant growth. PGPR include free-living nitrogen fixers or diazotrophs, such as Burkholderia sp., Azotobacter sp., Bacillus polymyxa and Azospirillum spp. [128]. Legumes with an active complement of rhizobia, AMF and PGPR produce organic acids that solubilize soil minerals and enhance the availability of rock phosphate [129] and mineral-based potassium [130]. These AMF and PGPR produce iron-chelating siderophores that both enhance plant uptake of micronutrients such as Zn, Cu and Fe [124] and suppress pathogenic microorganism development through iron deprivation [131].

These AMF and rhizobia can also cause plants to exhibit higher levels of photosynthesis, increased root growth, and enhanced plant production under stress conditions [132]. During drought or high saline conditions, AMF can produce aminocyclopropane-1-carboxylate (ACC) deaminase to protect plant roots and rhizobia from high levels of stress-induced ethylene [125,126]. Some PGPR exude bacterial exopolysaccharides (EPSs) [133,134] that protect against drought by maintaining water potential in plant roots. EPSs can also provide plant protection against diseases, facilitate PGPR contact with roots and enhance the formation of soil aggregates [135]. Soil aggregate formation, resulting in greater soil tilth, water and nutrient holding capacity and soil carbon sequestration, is a critical benefit of AMF growth [136].

Phytohormones produced by AMF, including salicylic and jasmonic acids, protect plants against pathogens by inducing plant defense mechanisms [137]. These defense mechanisms not only protect against soil-borne pests and pathogens but can also guard against predatory attacks on plant stems and leaves [138]. Some PGPR, including Psudomonous, Bacillus, Rhizobium and Microbacterium, produce auxins, such as indole-3-acetic acid (IAA), cytokinins, and gibberellins. These phytohormones enhance nutrient uptake and protect against pathogens by increasing root biomass, stimulating plant growth, decreasing plant stomata size and density and enhancing the release of plant root exudates [139,140]. These PGPR can also act as “mycorrhizal helper bacteria” to stimulate AMF inoculation and mycelial growth. Similarly, some AMF can facilitate inoculation of rhizobia and the growth of PGPR [124].

Rhizobia, AMF and PGPR evolved interactions with specific legume species under a given set of soil and climate conditions [124,141]. Successful field inoculation by potentially beneficial microorganisms is determined by biogeography, which affects adaptations at multiple levels, ranging from climate to micro-niche [142]. Benefits from these interrelationships appear to be based on specific combinations of rhizobia, PGPR and mycorrhizae as well as on the variety of host legume [143]. Legumes have evolved a suite of ‘host control’ traits that select for compatible rhizobia and defend against strains that provide little or no benefit [144]. AMF species can form associations with a variety of plant species and individual plants typically form associations with more than one species of AMF. Compatible AMF and host plants have high root infection and hyphal growth [145], resulting in high phosphorus uptake by plants with low carbon demands [146] and limited or short-term competition with rhizobia for nitrogen [124]. Mycorrhizae grown in native or “home” soil produce greater hyphal growth than mycorrhizae grown in non-native soil or soil with incompatible characteristics [147]. Risks to indigenous microbial communities can occur when inoculated non-native AMF varieties are more competitive than native AMF varieties [148]. Inoculation success with PGPR depends on soil nutrient concentrations, pH, organic matter content and temperature, which affect the ability of introduced PGPR to compete with existing soil microorganisms [149].

Efforts to translate the scientific understanding of the physiology and ecology of soil microorganisms into the commercial development of inoculants that enhance plant growth and resistance to stress conditions have produced variable results. Ineffective legume–plant relationships with AMF and PGPR due to incompatibilities between plants and microorganisms and competition between introduced and native soil microorganisms [150,151], as well as varying agricultural management practices, impact the populations and functional effectiveness of rhizobia, AMF and PGPR. Tillage [152] fertilization, pesticide use [153], clean fallows and crop rotations with non-mycorrhizal plants, such as Brassicaceae [154], impact microbial community structure, diversity and populations. Legume density, particularly in mixed fields such as pastures, can affect soil dissolved carbon availability and the activities of both rhizobia and diazotrophs, as well as their abundance and community composition [155].

Greater understanding of the biological, chemical and ecological factors affecting interactions between legumes and soil microorganisms, combined with ever-evolving analytical tools, enhance the potential for developing effective inoculants and inoculation practices in the future. Numerous methods are being investigated for creating effective inoculation products containing mutualistic microorganisms. Genetic sequencing can be used to select microorganisms (

Table 4. Soil microorganism contributions to soil and legume health.

	N Fixation	Enhance Nodules	Enhance AMF	P Solubilization	>K Uptake	Salinity and Drought Stress	Disease Suppression	Plant and Root Growth
Agrobacterium		X		X				X
Arthrobacter						X
Aspergillus						X
Azoarcus	X
Azotobacter	X			X		X		X
Azospirillum	X	X						X
Bacillus	X	X	x	X	X	X	x	X
Burkholderia	X				X		x
Diazotrophicus	X
Enterobacter			x	X		X		X
Herbaspirillum	X
Klebsiella	X			X				X
Pseudomonas		x	x	X	X	X	x	X
Rhizobium	X			X		X	x	X
Rhizopus						X
Serratia	X	x						X
Trichoderma							x	X

Sources: [124,126,140,142,162,163].

) with specific beneficial or environmental traits [156] and to identify processes influencing the establishment success and persistence in the soil of inoculants under field conditions [154]. Encasing individual and mixed cultures of beneficial microorganisms in microcapsules made of plant-based polymers enhances shelf life, allows for slow release of the microorganisms, and protects them from predation during the establishment stage [157,158]. To date, attempts at breeding selected plants and microorganisms for compatible symbiotic relationships currently appear to require organic or similar farming practices to be effective [137]. Alternatively, practices that rely on ecological processes to enhance crop production through microbial relationships include on-farm production of inoculants [159,160] and cover crops that share symbiont compatibility with the main crop [161].

7. Ecosystem Services

Ecosystem services are benefits ecosystems provide to entire human societies [164]. Forage legumes provide numerous benefits, including provisioning, supporting, regulating and cultural ES. Dubeux et al. (2017b) [3] and Sollenberger and Dubeux, Jr. (2022) [165] summarized different ES from forage legumes. Abiotic ES in grasslands derived from forage legumes include myriad non-agricultural benefits, resulting in more sustainable grasslands and encompassing the atmosphere and soil [165]. These include nutrient cycling, soil fertility/soil health, increases in soil organic matter, reduced N fertilizer surface and sub-surface water pollution, fewer greenhouse gas emissions and greater soil C and forage C accumulation and short-term sequestration. This reflects the potential of legumes, especially tree legumes, not only to increase overall productive potential, but also to curb GHG emission by enhancing C sequestration.

Biotic ES include additional biodiversity from herbaceous, brush and arboreal canopies. Direct biotic ES include browse and mast for browsing livestock ([59] Santos et al. 2019), timber and fuel [68], human food [166], natural medicines and ornamentals [167]. They also include biologically fixed atmospheric N₂, as previously discussed, that benefits microorganisms, plants and animals, especially enhanced photosynthesis and primary productivity, including C sequestration in plants and soil organic matter. Legumes in diverse systems can enhance overall primary ecosystem productivity and biodiversity by more efficiently using the natural resources in both space and time, due to greater species richness and the diversity of functional groups [168]. Apolinário et al. (2015) [68] reported an annual above-ground tree biomass accumulation of 10 Mg ha⁻¹, not considering the recycled biomass to the system via litter fall. These biotic provisioning ES are key to diversifying the income of livestock and cropping agroecosystems, resulting in sustainable operations.

Regulating ESs include C sequestration, GHG mitigation, soil erosion control, use of riparian buffers, shade and windbreaks and forage for pollinators. Net C sequestration occurs when ecosystems act as C sinks, meaning that C emissions are lower than C sequestration. Because livestock release methane, a powerful GHG, it is important to consider a life cycle analysis when assessing contrasting systems. Legumes might reduce GHG emission by producing condensed tannins, a plant secondary compound with potential to reduce rumen [169], fecal [170] or soil [171] methane release. Greater net primary productivity, especially in SPSs [68], coupled with products that have reduced C turnover (e.g., timber for construction and furniture), could further help to reduce C emissions and enhance overall C sequestration. Legumes could also recover eroded areas and prevent further ecosystem degradation by promoting a regeneration of the entire system [172]. Establishment of tree legumes along contour lines could also prevent soil erosion on sloping lands [173]. Grasslands with greater species richness, especially flower-rich vegetation, result in better habitats for pollinators [174]. Sowing legumes in grass monocultures typically enhances pollinator abundance [175].

Legumes can also have an important role delivering cultural ES, including recreational and aesthetic/spiritual values. In Florida (USA), it is common to find medians and urban landscaping of cities using legumes such as Arachis glabrata and A. pintoi. Likewise, native legumes such as Lupinus texensis are used as roadside beautification in Texas. Legumes are also important components of wildlife seed mixtures used to attract deer during the hunting season. Deer are browsers and have preference for forage legumes. Thus, native or seeded legumes have potential to diversify the diet of browsers [176], which have important value as cultural ES.

8. Economics

Legumes play a crucial role in the economic viability of major cropping systems around the globe. For example, in the USA, Glycine max planted in rotation with corn constitutes a major crop, where they confer improved long-term financial performance for Zea mays following G. max relative to continuous corn systems, especially in organic production rotations that incorporate M. sativa [177]. Overall, corn–soybean rotations tend to yield greater net incomes than continuous corn systems. The same is true of other grain–legume rotations. Forage legumes also improve gross margins of cropping systems.

As a major source of protein, legumes play an important role in human and livestock nutrition. From a food security standpoint, legumes are the primary source of protein in the feed rations of commercial livestock operations, and they remain the most prominent plant source of protein in human diets as well [178]. Legumes, primarily soybeans, are also a major contributor to biodiesel production, and various legume species are planted as cover crops for ecological and economic benefits. By enhancing soil nitrogen levels through microbial fixation, legumes also contribute to reducing greenhouse gas emissions by reducing nitrogen volatilization losses and fossil fuel utilization in fertilizer production.

8.1. Agroecosystem Profitability

The economic importance of legumes stems primarily from their capacity to fix N through BFN. Legumes, such as G. max, planted in rotation with cereal grains tend to confer greater cereal yields [179,180,181,182] and lower production costs [180,182] to the cereal crops, thus enhancing the overall profitability of the crop rotation [182,183]. Greater yields and lower production costs associated with cereal crops in rotation with legumes are due primarily, though not exclusively, to N fixation, which subsequently becomes available to the cereal crop. In Europe, legumes reduce nitrogen fertilizer use by 24% and 38% in arable and forage systems [184]. Consequently, cereal crops in rotation with legumes are often associated with greater profits than when continuously cropped. This is often true [185,186] even if, on average, the cereal crop itself is more profitable than the legume [186] Dobbins and Langemeier 2018). Furthermore, for organic farming systems that rely on non-synthetic fertilizer sources, legume crops represent an affordable alternative to expensive organic nitrogen sources.

Besides their contribution to soil nitrogen levels for the benefit of the subsequent crop, legumes used in rotation also help to reduce the incidence of pests, thereby limiting the need to apply pesticides. As a break crop, they disrupt the cycle of weeds and other pests that thrive in mono-cropping systems [187]. Legumes grown for forage [188] or bioenergy [189] also provide substantial benefits to grass pastures, bioenergy or subsequent grains. In southern North America, for example, annual cool-season legumes (e.g., Medicago or Trifolium spp.) are often overseeded onto dormant perennial warm-season grasses to enhance soil nutrient levels for the subsequent grazing season. In addition, some perennial forage legumes (e.g., M. sativa) are grown by themselves as a major source of nutrition for dairy cattle that require a high protein intake to support profitable milk or meat production.

8.2. Economic Flexibility and Risk Mitigation

Farmers routinely deal with one or more of five major types of risk: production risk, market risk, institutional risk, personal risk and financial risk [190]. Among these, production and market risk are two major risks that can be managed through enterprise diversification. Incorporating legumes in a production system helps to mitigate risks associated with the production process and market forces.

Production risks that legumes can mitigate relate to the inherent variability in output that stems from exogenous factors such as weather events. Several aspects of production risk are mitigated by incorporating legumes in a cropping system. First, there are risks associated with the timing of N fertilizer applications [191]. Unforeseen precipitation events may make it impracticable to apply nitrogen fertilizer at the most appropriate time, resulting in lower than optimal nutrient levels for non-legume crops. The production risk associated with such situations is a function of stochastic weather events. In such instances, legume pre-crop benefits ensure N availability to the crop without the need for an additional N application operation [187]. In addition, N volatilization, surface runoff or soil leaching are mitigated. All these losses are minimized when the preceding crop is a legume, resulting in more stable yields of the non-legume crop.

A second aspect of production risk relates to pest pressure associated with mono-cropping. Legumes can serve as break crops to disrupt pest cycles that rely on consistent mono-cropping patterns to maintain their habitat. Break crops may reduce pest pressure, thus resulting in reduced volatility of year-to-year variations in the yield of the non-legume crop [187].

Besides the pre-crop benefits legumes confer to non-legume crops, a more fundamental benefit is that the legume crop itself provides a product the producer can rely on for additional income or livelihood. Furthermore, many legumes are known to generate multiple products or by-products that can also provide additional income for farmers. For instance, a single G. max crop can generate several by-products including soy meal, flour and oil, among others. The by-products may be (such as A. hypogea peanut butter and whole peanuts) or may not be (such as soy meal and oil) mutually exclusive. In either case, farmers have more certainty of an economically beneficial crop if multiple potential uses are available from one legume harvest.

Market risk relates to the uncertainties in prices, costs and market access that stem from domestic or global market forces. Price volatility is perhaps the most obvious component of market risk. Price changes have a direct impact on farm profits. Consequently, price risk is mitigated by diversifying an enterprise to include crops or livestock whose prices are countercyclical to the price of the initial crop or livestock, and legumes are not unique in their role in this respect [187]. Any crop or livestock whose price is negatively correlated with the existing crop or livestock can be added to minimize price risk. The greater the diversity of products or by-products entailed in the enterprise, the more options the producer will have to mitigate market risk.

Market access is a source of market risk that is potentially heightened by global market forces. For instance, the incidence of African swine fever in Asia in recent years has led to massive culling of swine herds in Asia, which is expected to have negative ramifications for U.S. G. max exports to Asia [192]. As with production risk, a broader portfolio of crop or livestock by-products will help to mitigate risks that stem from market access.

8.3. Specific Contributions

Legumes as cover crops provide significant ecological and economic benefits to commercial crop and livestock enterprises. Cover crops suppress weeds, insects and pathogens. Pest suppression offers economic benefits through reduced pesticide use [193] in many instances. In addition, cover crops improve soil health and reduce direct loss of soil by wind or water erosion. Hansen and Ribaudo (2008) [194] estimate that, in the USA, each ton of soil retained through conservation practice implementation is worth between USD 1.70/year to USD 18.24/year depending on the biophysical attributes and management practices implemented on the land. Thus, as the erosion reduction benefits of cover crops increase over time with improving management skills, the overall benefits of cover crops increase until the maximum benefit is obtained (Figure 1; adapted from [193]).

In addition to the standard benefits of cover crops in general, leguminous cover crops provide N immobilization [194]. The direct economic benefit of BFN from cover crops increases with the price of N fertilizers. At the current N fertilizer price of USD 0.88/kg [195], various legume cover crops confer an economic value of between USD 35 and USD 106/ha depending on how the cover crop is managed (

Table 5. Fertilizer cost savings from green manure N credits of legume cover crops (USD/ha/year) *.

Cover Crop	<15 cm Growth	15 to 30 cm Growth
Alfalfa	USD 35	USD 53–USD 88
Red clover	USD 35	USD 44–USD 70
Sweet clover	USD 35	USD 70–USD 106
Vetch	USD 35	USD 35–USD 79

* Adapted from [197], based on a fertilizer price of USD 0.88/kg.

). The economic benefit depends also on the portion of the immobilized N that is taken up by the subsequent crop [196].

Notwithstanding the aforementioned benefits, cover crops entail additional costs that are not otherwise incurred by crop and livestock enterprises. Seeding and management of the cover crop entail additional expenses, including initial capital outlays that decrease over time, and regular operating and maintenance expenses that are incurred in each season. Conversely, as indicated above, the soil productivity benefits mentioned above improve over time. Thus, in many instances, the costs may outweigh the benefits during the initial years after implementation. Ultimately, the cover crop becomes economically viable as increasing levels of benefit outweigh annualized costs that decrease over time (Figure 1; adapted from [193]). Legume cover crops confer additional benefits in the form of reduced N fertilizer costs, so the payback period—the time required for cumulative benefits to outweigh cumulative costs—is lower than with non-leguminous cover crops. The dynamics of costs and benefits indicate that financial incentives may be necessary to encourage farmers to adopt cover crops on their operations. Once the payback period is reached, management of the cover crop will produce benefits that exceed the cumulative costs, and no additional monetary incentives would be needed.

Around the world, legumes are a significant source of nutrition. While grain legumes are not generally classified as a complete protein source, they are among the most concentrated sources of protein of any food material [198]. Grain legume products and forage legumes are primary sources of protein in the rations of commercial livestock operations. Since livestock rations are largely developed using feed ration optimization algorithms, the prevalence of grain legume by-products such as soybean meal in livestock rations implies that they are among the most affordable sources of protein available for raising livestock for human consumption. Consequently, through consumption of animal protein, humans in turn rely heavily on legumes as a major protein source. Diets that are based on non-animal protein sources are even more heavily dependent upon legumes—primarily nuts and beans—as the main source of protein. While the United States of America is a net exporter of plant and animal protein, the European Union, where legumes cover only 2% of the cultivated land area, imports most of their required protein from other regions [185].

Historical data from the Energy Information Administration indicate that soybean oil is consistently the most prominent source of feedstock for biodiesel production in the USA [199]. As of December 2019, soybean accounted for about 50% of biodiesel feedstock in the U.S., and annual data from the Energy Information Administration suggest that this has been relatively consistent in recent years (Figure 2; [200]). Approximately 1 kg of soybeans generates 1 L of biodiesel [201]. Thus, given current average G. max yields of 3230 kg/ha [201], each hectare of harvestable soybeans has the potential to contribute 630 L of biodiesel. Glycine max oil as a bioenergy feedstock has contributed an additional revenue stream that supports grain soybean prices for the benefit of farmers [202]. Soybean oil is a joint product in the soybean crush process that generates soybean meal as a main product for the livestock feed industry. The Renewable Fuel Standard (RFS) in the USA, along with increases in soybean imports by emerging economies, resulted in a recent economic boon for soybean farmers. These policy and market forces have enabled soybean farmers to enjoy higher prices while maintaining high yields on greater acreage than in previous decades.

8.4. Policy Challenges and Barriers to Adoption

The foregoing discussions indicate clearly that legumes play a crucial role in sustainable agricultural systems. This is especially true when one considers the role forage legumes play in mitigating the effects of climate variability and change. In particular, forage legumes provide a more sustainable contribution of nitrogen to the soil than conventional high energy nitrogen fertilizer production processes.

These benefits notwithstanding, the current incentive structure of many agricultural policies, at least in the U.S., has not been favorable to incorporating forage legumes in livestock production [203]. The primary reason for this is because farm policies—particularly subsidy programs—have favored high-input, high-yield production systems primarily tailored to row crops, sugar, peanuts, wool, tobacco and dairy products [204]. For this reason, farmers are incentivized to produce commodities such as corn, sugar and wheat that are subsidized at the expense of those that are not.

There are many other reasons why farmers tend to prefer high-input, high-output production systems. One reason is that in making input decisions, many farmers tend to rely on crop advisors and technical experts that have a vested interest in selling chemical inputs as a yield insurance mechanism. Another reason for the lower incentive to incorporate forage legumes is because they are not covered by major farm insurance programs. Many of these programs also encourage high-input, high-yield production, since insurance payments in the event of yield losses are often pegged to historical yield averages. Yet another reason for the difficulty in incentivizing forage legume production is due to the fact that conventional large-scale mechanized production systems tend to be more economically viable in the short run, as compared to forage legume systems that are often raised at a much smaller scale.

For all these reasons, it is difficult to incentivize adoption of long-term sustainable systems that incorporate forage legumes. To foster a more sustainable landscape in agriculture, farm policies and programs need to be tailored to provide more equitable attention to low-input systems, including forage legumes and various mixed production systems, that offer more support for mitigating extreme climate events [205].

9. Genomics

Although farmers have improved legume versatility for millennia without any knowledge about the genetic mechanisms of inheritance, technological advances in the study of DNA and RNA position plant scientists to speed legume improvement with greater design and intent in order to keep pace with the increasing human demand noted above. While genomic tools now enable legume breeders to detect and advance many quantitative trait loci (QTL) at once in breeding populations, initial efforts in legume genetics involved single genes or markers linked to them. Early successes included identification of single genes like those in Gregor Mendel’s segregating peas [206]. Technological advancements in DNA sequencing enabled progression from markers to genes to genomes. Following on the heels of the Human Genome Project [207] and the first plant genome sequencing effort for Arabidopsis thaliana [208], the first legume whole genome sequencing efforts in Lotus japonicus and soybean, chosen for their small genome size and economic importance, respectively, Refs. [209,210] utilized methodologies developed in those pioneering whole-genome sequencing projects, and involved large consortia of scientists pooling resources to accomplish the herculean task, progressing from genetic mapping and physical mapping to sequencing and assembly of draft genomes via Sanger sequencing [211,212]

Technological advances in DNA sequencing, and equally important, advances in analyzing vast numbers of DNA sequences, have transformed what is possible for legume research. Due to next-generation sequencing or massively parallel sequencing, it is now possible for a small group of scientists or even a single research group to produce a draft legume genome sequence, or reduced representation genome sequences from hundreds of individual plants, for marker generation and genome-wide association studies (GWAS). Many university core laboratories and commercial sequencing laboratories conduct fee-for-service sequencing, starting with plant material with nucleic acid extractions, or with fully prepared sequencing libraries, depending on the needs of the researcher. Importantly, bioinformatics support is also readily available at many sequencing facilities to advise and assist in project design and analysis as a fee-for-service.

Among the DNA sequencing technologies now available, short-read Illumina sequencing currently dominates the sequencing market due to its high read quality, high output, and low cost, with read lengths varying from 50 to 300 base pairs, while another short-read sequencing technology from Ion Torrent is still in use. The sequencing costs and output for Illumina sequencing are scalable depending on the sequencing platform. Two types of sequencing experiments that offer high return-on-investment using short-read technology, particularly for researchers employing genomics approaches for the first time or working on orphan legume crops, are reduced representation sequencing and transcriptomics.

Reduced representation sequencing methods sequence only a portion of the genome. Reduced representation sequencing, and any of several similar methodologies exemplifying reduced representation sequencing, involve sequencing only the portions of a genome adjacent to DNA breaking points at restriction sites or other genomic motifs [213,214,215,216]. Because the restriction sites are distributed throughout a genome, reduced representation sequencing enables marker generation across a genome without any a priori knowledge of the genome sequence or structure. Reduced representation sequencing can produce tens of thousands of polymorphic single nucleotide polymorphism (SNP) markers in a single experiment, saturating a legume genome with markers. When combined with phenotypic data from individual plants, the markers provide associations between SNP markers and traits of interest. Importantly, while DNA sequencing throughput has increased, sequencing costs have rapidly declined. A reduced representation sequencing experiment can now be conducted on 96 samples for a cost in the neighborhood of USD 4000 for library construction and sequencing. Reduced representation sequencing methods have proven useful for exploring the phylogenetics of legume species as well as for identifying QTLs of economic importance [217,218]. The bioinformatics for a reduced representation sequencing effort can typically be handled by a single individual.

Another genomic method that can produce high return on investment is analysis of the transcriptome—all of the genes being transcribed into RNA messages at a point in time. While more expensive than reduced representation sequencing (now about USD 10 K for 32 samples), transcriptome sequencing provides a wealth of targeted information on the plant response to the environment. For instance, one might compare gene expression of a legume genotype in well-watered and drought conditions, or compare between drought-resistant and drought-susceptible genotypes. The imposition of any biotic or abiotic perturbation may lead to gene expression differences that can be documented in a transcriptomics experiment. Genes contributing to trait differences are identified by their increased or decreased expression in response to a stimulus [219]. Polymorphisms identified between the alleles in differing genotypes can additionally serve as DNA markers for use in plant breeding [220]. Finally, the bioinformatics for a transcriptomics experiment can also typically be handled by a single scientist.

Larger genomics projects, such as whole genome sequencing, are now possible for small research groups. While short-read Illumina sequencing can produce more than enough genome coverage for whole genome sequencing at a very modest cost, the high repetitive element content in legume genomes and the polyploid structure of some legume genomes create problems when assembling short sequencing reads into a draft genome sequence. Therefore, whole genome sequencing efforts often involve a combination of Illumina sequencing and long-read sequencing technologies (such as Pacific Biosystems and Oxford Nanopore) that provide lower read quality but much longer reads, facilitating assembly of more complete draft genomic sequences, or modifications of short-read Illumina sequencing that provide much longer-range genomic information (10X Genomics, Pleasanton, CA, USA) [221]. As long-read sequencing matures, it is being used as the primary technology for whole genome sequencing [222]. Additional novel tools such as optical mapping and chromosome conformation capture may be utilized to resolve chromosomal rearrangements [223,224]. The bioinformatics for whole genome assembly in the absence of a previously sequenced reference genome is more demanding than for reduced representation sequencing or transcriptomics. The cost and difficulty for creating and assembling a legume whole genome sequence depends on the genome size and complexity (autopolyploid > allopolyploid > diploid and hexaploid > tetraploid > diploid) [225]. As a reference, an allopolyploid legume with a genome size of about 4000 Mbp can now be sequenced and the draft assembled for approximately USD 50 K using long-read technology alone. Bioinformatics for whole genome sequencing can be more complicated, as reflected in the higher cost of a whole genome sequencing effort, but a reference genome sequence for a species can greatly facilitate analysis of reduced representation sequencing, transcriptomics, and legume breeding efforts in general.

As genomic information accumulates for a legume species, or closely related legumes, and a multitude of SNPs are linked to chromosomal loci, GWAS studies are possible. GWAS studies can identify genetic variants that influence yield parameters, biotic/abiotic stress responses, beneficial microbial interactions and nutritional factors [226,227]. At the level of GWAS, the tools of genomics are identifying genes and alleles that control complex genetic traits, providing insights into legume biology that can be used to develop comprehensive breeding/improvement strategies.

Unlike novel/underfunded/orphan legume species, legume species of high agricultural importance now have a vast array of genomic tools and information readily available, including multiple annotated reference genomes of important cultivars and close relatives enabling skim-based genotyping, SNP arrays for inexpensive and rapid genotyping, reduced representation sequences from populations segregating for many traits of importance, and online databases for data curation and comparative genomics. Advances in legume genomics have been summarized recently [228,229,230], including methodological developments. The last decade has seen a rapid change in DNA sequencing cost, quality and scale. Some massively parallel sequencing technologies in use a few years ago have already disappeared (Pyrosequencing and SOLiD sequencing), having been rapidly replaced by the current sequencing methods. The falling cost of sequencing is democratizing genomic studies, enabling even modestly funded research groups to participate. When combined with technological advancements in phenomics, proteomics, metabolomics and plant transformation, the coming decade holds great promise for breakthroughs that will speed the improvement of legumes, even for species making niche contributions to human needs. Recent revolutionary changes in DNA sequencing methods are contributing much to understanding the genetic basis of important legume traits. That understanding will fuel legume improvement for decades to come.

Once genes influencing legume traits have been identified, either by DNA markers linked in close proximity to the gene or by identification of the gene itself, legume breeders can use multiple methods to improve existing lines, depending on circumstances. If the gene of interest is in a wild legume relative that is capable of hybridizing with elite legume materials, the gene of interest can be introgressed into elite legume lines by traditional plant breeding methods. These methods involve crossing the donor to elite lines, followed by multiple backcrosses of offspring to the elite material to remove the unwanted “wild” portions of the donor genome while keeping only the desired portion of the donor genome contributing the gene of interest. The gene of interest can be tracked by phenotype or by DNA markers if the phenotype is not easily confirmed. In circumstances where the gene of interest cannot be easily introgressed into elite germplasm, a recently developed targeted DNA insertion technology called CRISPR (Clustered Regularly Interspaced Short Palindromic Sequences) may be utilized to insert DNA in a genome. CRISPR is a bacterial nucleic acid defense system that has been modified and repurposed as a genetic engineering system. In legumes, as in other organisms, the CRISPR system is currently being used to insert genes with positive impacts and delete genes with negative impacts with far greater ease and efficiency than ever before, allowing improvement of diverse legume species [231,232].

Between the combination of our vastly increased ability to describe the genetic basis for important legume traits provided by novel nucleic acid sequencing methods and our ability to modify the DNA sequence at targeted loci through conventional breeding and site-specific insertion, our capacity to use targeted methods for legume genetic improvement has increased dramatically over the past two decades. The novel tools and massive decrease in costs for genomic studies have opened up possibilities for genome-scale studies and genomic improvements in any legume, even for modestly sized/modestly funded research programs.

10. Conclusions: Versatility into the Future

Versatile legumes already play an essential role in sustainable agroecosystems; we predict that they will make even greater contributions into the future. They have myriad functions: human nutrition, animal feed, ES, fossil fuel-free bioenergy and construction material, among others. Public research support, for example, at national research institutes and universities, will be essential to fully harness that versatility, especially in novel species that are currently unknown or regional. Educating and supporting private investment and cultivation will likewise be essential. In both the public and private sectors, expanded genetic diversity in currently exploited species as well as exploring novel germplasm will multiply legume benefits.

Forage legumes are not a panacea. There are limitations to their utility and in no way will they ever replace grasses in providing energy for human consumption (grain) or ruminant nutrition (forage). In addition, they sometimes have other limitations, such as slower establishment and less erosion control vis-à-vis grasses. However, we have enumerated the many ways in which their use can enhance agroecosystem sustainability.

Multiple uses contribute to forage legume versatility. A crop that can simultaneously provide pulses for direct human consumption, pharmaceuticals to keep humans or their crops healthy, fuel to cook those pulses, BFN to the soil to grow non-legume grain, forage for a milk goat and shade for the farmer’s house is the ultimate versatile domesticated legume. Such species exist, as exemplified by Leucaena leucocephala [233,234,235,236]. This and other Leucaena species also offer adaptability to varying edaphoclimatic conditions and a versatile genome that plant breeders can manipulate to meet changing market or abiotic conditions. If we compile a list or add new species around the world and harness their potential, we have our versatile legumes.

Germplasm exists to support greater use of versatile legumes. We have the tools to maximize our understanding and application of that genetic variability. If we harness that versatility in widely used species, as well as explore novel or historically underutilized legumes, we will expand legume applications and functions within current or future environments under biotic, abiotic and human-induced stresses. Diversifying pulse, forage or forestry uses within widely used domesticated species but also in under-utilized or novel genera and species could uncover applications that will stabilize agroecosystems into the future, regardless of what lies ahead. Harnessing legume versatility to a greater extent into the future will also stabilize agricultural ecosystems as these adapt to human pressures, volatile climates and soils depleted by non-BFN species. Climate or market changes will further favor versatile legumes adaptable to new short-, medium- and long-term stressors.