When combined with the state’s $10.4B value of milk production, alfalfa accounts for a significant portion of California’s agricultural GDP. In addition to its advantages as a forage, alfalfa also provides a host of beneficial ecosystem services: as a legume, alfalfa requires no nitrogen fertiliser, instead it fixes atmospheric nitrogen through a symbiotic relationship with rhizobia, a nitrogen fixing bacterium; the perennial nature of alfalfa improves soil health allowing fields to recover from frequent tilling and prevents topsoil loss; long roots can access water and nutrients deep in the soil profile and increase soil organic matter throughout; and alfalfa is an important insectary, hosting a diversity of beneficial insects . Yield is the most important trait for profitable alfalfa production, yet somewhat inexplicably, yield improvement in alfalfa has stalled over the last ~30 years . In addition to the lack of yield improvement and despite the significant economic and environmental benefits of alfalfa, the area cultivated has been in steady decline both nationally and in California since a peak in 1960 . Already disadvantaged by the distribution of federal subsidies, of which commodity row crops receive billions each year, alfalfa faces stiff competition for inclusion in crop rotations. Increasing yield is therefore imperative to curb the rate of decline in alfalfa area, not only to continue to support California’s significant livestock industry, but also for the benefits alfalfa provides to agricultural ecosystems. Historically, breeders have relied on traditional breeding methods to increase biomass yield in alfalfa, with little success in recent years . However, black flower buckets with the advent of new breeding technologies, such as genomic selection and high throughput phenotyping, this lack of yield gain may be approached in a new light.
This PhD project investigates the incorporation of modern breeding tools and methodologies into an existing breeding program to address the lack of yield improvement in alfalfa.Alfalfa is believed to have two centers of origin, Asia Minor/Caucasia and Central Asia . Following domestication, alfalfa quickly spread throughout the ancient world due to its adaptability to diverse climates and soils. The expansion of civilizations and the Silk Road trade routes played a pivotal role in the dissemination of alfalfa, enabling its introduction to regions such as Europe, North Africa, and eventually the Americas. Alfalfa exists naturally as the Medicago sativa complex that includes a range of diploid and autotetraploid subspecific taxa . Cultivated alfalfa primarily refers to the subspecies sativa and is an autotetraploid , perennial, outcrossing legume with polysomic inheritance . It has a basic chromosome number of eight and a genome size of 800-1000 Mb. Commercial alfalfa stands typically last three to eight years depending on variety, soil, climate, and cultural practices. Pure stands are sown at high density with typical sowing rates ranging from 16-22 kg ha-1 . Plant density starts high with up to 800 plants m-2 three months after planting and then steadily declines. In established alfalfa stands, plant density has limited effect on dry matter yield per hectare until plant numbers fall below 40 plants m-2 , wherein yields decrease and growers should consider retiring the stand. Alfalfa can grow to heights above one metre and has a deep root system that reaches beyond six metres in depth when grown in deep, well drained, moist soils . Alfalfa plants generally have a single deep taproot, with variation in the number and size of lateral roots. Following establishment, alfalfa forms a crown at the top of the root system. After defoliation, alfalfa regrowth occurs from buds located on the crown and from axillary buds at nodes from the remaining above ground stubble . This regrowth cycle allows for multiple cuttings during the growing season for up to eight years before forage yields decline below economic thresholds.
Fall dormancy is an important characteristic of alfalfa that defines a population’s fitness for specific agricultural environments. Fall dormancy is a plant’s response to decreasing photoperiod and temperature and is associated with a slowing and eventual cessation of growth through the dormant period . This trait is closely related to the ability for populations to avoid winter kill and is under complex quantitative genetic control . Alfalfa varieties are classified into different fall dormancy groups, typically ranging from 1 to 11, with lower numbers referring to dormant varieties and higher numbers representing reduced fall dormancy. Non-dormant varieties tend to be higher yielding, but may be lower quality and less persistent, therefore appropriate variety selection is of key concern to the grower.Californian agricultural regions have a broad range of climates and soils and can be divided into five main growing regions: the Central Valley, Intermountain, Low Desert, High Desert, and Coastal regions . Statewide average dry matter yields for alfalfa are 15-17 MG ha-1 from an average of 6-7 harvests per year although this varies with up to 12 cuts on stands grown in the Low Desert and as few as 3 on fields in the Intermountain region . The Central Valley is the most significant area accounting for 70 percent of the state’s alfalfa production . It has a Mediterranean climate characterized by hot, dry summers and cool winters . Rainfall totals range from 20-46 cm annually, falling predominantly in the cooler months from November through to March. The Central Valley has fertile, deep, alluvial soils, although some areas suffer from high salinity. Varieties grown in the Central Valley are predominantly semi-dormant to nondormant however, in the northern tip of the valley dormant varieties are grown for better persistence in heavy soils and greater forage quality. The Low Desert region in Southern California contains 17 percent of alfalfa production.
An area with extremely low rainfall and high temperatures . Soils are generally heavy and like in the Central Valley, high salinity is a significant issue. The Low Desert is also an important alfalfa seed production area. The final area of significance for alfalfa production is the Intermountain region in Northern California accounting for approximately 10 percent of production . This area is more temperate, located at high elevation with warm summers and cool winters . Rainfall averages about 51 cm per year falling in the cooler months. Freezing winters necessitate the use of dormant cultivars to prevent winter kill. Almost all the alfalfa grown in California is irrigated. Check-flood surface irrigation is the most common in the Central Valley and Low Desert regions, while sprinklers are the preferred method in the Intermountain area. Although a significant crop in California, the area of land cultivated in alfalfa has been in steady decline since a peak in 1960. Over the last 12 years hectarage has decreased by more than 40 percent from 390,000 ha in 2011 to 235,000 ha in 2022 . The predominant end-use of alfalfa grown in California is hay for dairy cattle. California surpassed Wisconsin as the number-one dairy state in 1993 and now produces more than 21 percent of milk in the United States . At least 75 percent of alfalfa grown in California is used to supply the dairy industry.The breeding goals for alfalfa are characteristic of most plant improvement programs. They include increasing yield, enhancing nutritive value, and improving biotic and abiotic stress tolerance . The majority of desired traits are complex and quantitatively inherited; however, some pest and disease resistance mechanisms are likely under simple genetic control.Alfalfa breeding programs are based on recurrent phenotypic selection, with or without progeny testing. They are designed to increase the frequency of desirable alleles for quantitatively inherited traits, while maintaining genetic variability for continued genetic improvement . Although self-fertilization is common, alfalfa suffers from severe inbreeding depression which is prohibitive to the production of hybrids, french flower buckets thus commercial cultivars are marketed as synthetic populations generated by crossing different numbers of selected genotypes . As a consequence of the structure of the alfalfa genome, cross pollination and severe inbreeding depression, cultivars exhibit high levels of genetic variation . Significant gains have been made in most traits of interest in alfalfa. Forage quality has improved, demonstrated by the release of new high-quality cultivars . Current cultivars exhibit resistance to a suite of pests and diseases including bacterial wilt , Verticillium wilt , Fusarium wilt , anthracnose , Phytophthora root rot , Aphanomyces root rot , root-knot nematodes , stem nematode , spotted alfalfa aphid , pea aphid , and others . However, there has been little to no improvement in yield, for which a variety of explanations have been proposed. Perennial forage breeders are at a disadvantage compared to those working with annual grain crops when it comes to yield improvement due in part to its perennial nature requiring multiple years of evaluation before selection can be made, the negative genetic correlation between forage quality and forage yield, and the inability to make gains in the harvest index that is possible in grain crops, as all above ground biomass is harvested . A lack of forage breeders and limited resources constrain the size and scope of breeding trials, reducing their efficacy in improving traits with low heritability, such as yield.
The emphasis by alfalfa breeders on pest and disease resistance and greater persistence may lead to a realization of yield potential but it is not increasing yield per se. Perhaps the most important reason for the lack of yield improvement in alfalfa is that breeders have not been explicitly selecting for increased yield potential under commercial production conditions. Yield is often selected indirectly based on evaluation of vigor on spaced plants or on short family rows rather than on measurements of yield on plots grown as a densely sown sward . Although little can be done to address issues such as the inability to increase the harvest index of alfalfa, we do have the ability to modify our trial methodology and to utilize modern technology such as genomic selection and high throughput phenotyping to improve yield potential in alfalfa. Measuring yield on plots instead of individual plants provides a better proxy of commercial production systems, and the incorporation of remote sensing, high throughput phenotyping and genomic selection allow for better allocation of resources within a breeding program and can help to speed up the rate of genetic gain through shorter selection cycles.Genomic selection is a form of marker-assisted selection where the breeding value for a given trait of a genotype can be estimated from many markers distributed across the entire genome . MAS is based upon the establishment of a tight linkage between a molecular marker and the chromosomal location of the gene governing the trait to be selected in a particular environment . This is useful when working with traits controlled by a few large-effect loci, however this is not the case for many important traits in plant breeding, including biomass yield. GS expands on this idea and guides selection based on the cumulative impact of many small-effect loci that are in linkage disequilibrium with genetic markers . To estimate breeding values for a genotype, first a model must be developed from a training population that exhibits variation for the trait of interest. This training population is both genotyped and phenotyped to develop a model which optimizes the chromosomal pattern of alleles. This model then allows breeders to estimate the genetic potential of unobserved genotypes based solely on marker information. Because it relies only on a plants genotype, which can be determined when the plant is still very young, genomic selection allows significantly shorter selection cycles than are required under recurrent phenotypic selection, thus has the potential to significantly increase the rate of genetic gain in an alfalfa breeding program. Early methods of GS in plant breeding used gene chip arrays and were mostly adopted by well-resourced breeding programs in major crop species . As high-throughput sequencing costs have decreased, sequencing based approaches have become more viable and GS is now being explored in a wide range of crops. Genotyping-by-sequencing is currently the most popular reduced-representation approach for genomic selection . It is a reduced-representation approach that uses restriction enzymes to cut the genome at fixed points to show good overall coverage. Samples are sequenced using high throughput next-generation sequencing platforms. A variety of software programs can then be used for genotype calling and identification of polymorphisms for downstream analysis . Because genome coverage is incomplete, causal loci are unlikely to be identified, but the polymorphisms are likely to be in linkage disequilibrium with loci that are causal.