Determining the performance of alfalfa population collected from a narrow agroeceological zone of Turkey

This study was carried out to determine the yield and quality characteristics of some alfalfa village populations cultivated by producers in Bingol province of Turkey. Alfalfa seeds were obtained from 23 different locations in 2015. A three replicated, randomised complete block designed field trial was established in 2016 including these genotypes and four registered varieties. As a result of three year trials, it was determined that the populations obtained from the villages of Servi, Mutluca, Garip, Sarıçiçek-1 and Bagliisa produced highest green herbage and hay yields. When we compare the quality results, Kumgecit, Küçüktekören and Şenköy-1 populations stands out in terms of crude protein ratio. Highest crude protein yield was obtained from Servi population. In terms of relative feed value, it was observed that the varieties were better qualfied than the populations. It is concluded that the populations of Servi, Mutluca, Garip, Sarıçiçek-1 and Bagliisa can be used in breeding studies especially to obtain genetic progress in yield of alfalfa for feed and bioenergy production. For the genetic improvement of current genotypes with higher crude protein ratio, Kumgecit, Küçüktekören and Şenköy-1 populations can be used as a source. This study showed that, East Anatolia region is not just live-conserving highly diversified species but also covering special intra-species genetic diversity in microclimatic zones of Turkey to be used to improve the global forage and bioenergy cropproduction.


INTRODUCTION
Alfalfa (Medicago sativa L.) is a widely cropped legume forage for hay, pasture and silage production throughout the World (LI & BRUMMER, 2012). It provides significant habitat to a wide range of wildlife, including beneficial insects. It improves yields and reduces nitrogen fertilizer needs for subsequent crops. Alfalfa also helps to remediate soil and water contamination (PUTNAM et al., 2001).
Alfalfa also has considerable potential as a feedstock for production of ethanol due to its high biomass production, perennial nature, ability to provide its own nitrogen fertilizer and valuable co-products. Its stems are an excellent feedstock for cellulosic ethanol via fermentation or gasification (MARTIN et al., 2010). The stems has a potential as biomass energy crops especially in temperate regions where their bioconversion potential is high as an energy crop (DIEN et al., 2006). Alfalfa has been considered as a biofuel feedstock where a system has been proposed to produce electricity from the stems following utilization of the leaves as a livestock feed (SHEAFFER, 2000).
Increasing forage yield is main priority for most alfalfa breeding programs, but yield trends has stagnated over the past two decades (RIDAY & BRUMMER, 2002). Currently, alfalfa breeding relies on recurrent phenotypic selection (LI & BRUMMER, 2012). Alfalfa cultivars are synthetic populations formed from 8 to 200 parents and have a broad genetic base. In the study of JULIER et al. (2000), when intra-cultivar variation was compared with amongcultivars for energy value traits, morphological traits and dry matter yield, variance accounted for 31 to 70% of the genetic variance for leaf-to-stem ratio and quality traits and 57 to 100% for morphological traits and dry matter yield. Researches informed that large intracultivar variation for yield-related traits could impart yield stability across environments by competition in alfalfa canopies and cultivar variation could be an additional source of genetic variation in breeding programs for quality traits to achieve a higher genetic gain per breeding cycle (JULIER et al., 2000).
The most cultivated feed crop in Turkey is also alfalfa. Since this species is perrennial, well adapted to different conditions, moweable multiple times in a single year vegetation period, good rotation crop, in between with its high hay yield and quality (SoYA et al., 2004;AVCIoGLU et al., 2009). SARUHAN & KUSVURAN, (2011) determined that the yield performance of some alfalfa genotypes in Southeast Anatolia Region conditions of Turkey was at good levels. They determined green grass yield at 35.1-19.0 t/ha, hay yield at 9.4-12.6 t/ha, the ratio of crude protein at 17.9-22.3%. To determine the yield of some alfalfa genotypes in Mediterranean climatic conditions in Turkey, KAVUT et al. (2014) carried out a field trial and obtained 95.2-118.9 t/ha green grass yield, and 20.3-27.1 t/ha hay yield. In the study of oTEN & ALBAYRAK (2018), which examined the quality characteristics of 26 alfalfa genotypes collected from South-West natural flora of Anatolia, the average crude protein ratio was 15.8%, the ADF (acid detergent insoluble fiber) rate was 36.7% and the NDF (neutral detergent insoluble fiber) rate was 47.2%.
In other important grass production zone of Turkey, in East Anatolia, alfalfa is covering 7.392 ha solely in Bingol province, with 292.000 tons production and 39.5 t/ha green grass yield (TURKSTAT, 2019). Alfalfa hay yield differs among regions in Turkey where is is 2.5-3.0 t/ha in rainfed, 5.0-25.0 t/ha in irrigated conditions (SoYA et al., 2004;AVCIoGLU et al., 2009). Since the main source of income in Bingol province of Turkey is animal husbandry, alfalfa cultivation has a long history. Alfalfa cultivation in Bingol is mostly carried out by population seeds.
The purpose of this study was to determine the yield and quality characteristics of some alfalfa populations in Bingol with the aim to utilise them in breeding programmes.

MATERIALS AND METHODS
Research field was located in Bingöl province, Genc subprovince of Turkey. Altitude of the field was 986 m and located at 38.749460 North latitude and 40.536780 East longitude coordinates. In 2015, seeds of local alfalfa populations were obtained from farmers in the villages of Bingol. A total of 23 local genotypes were collected.
The trials was conducted in Bingol University research fields. The research area is located at 38.749460 North latitude and 40.536780 East longitude coordinates (Figure 1). Long term  average temperature is 12.3 o C, precipitation is 917.8 mm and humidity is 56.6% in Bingol. July and August are the months with highest temperature with lowest rainfall and humidity. Most of the precipitation realises in winter period (Table 1).
Long yers  average climatic records obtained from Bingöl Provincial Meteorological Centerindicates that average air temperature of the trial location is 12.1 o C and precipitation is 943.3 mm (ANoNYMoUS, 2019).
Soil samples obtained from trial field were analysed in University of Bingol Soil Analysis Laboratory. For the evaluation of the analysis results, limit values determined by of SEZEN (1995) was used as base. Soil of research area is sandy-clayey-loam structured (59.5% sand, 18.2% clay, 22.3% loam), has neutral pH ratio (7.26), medium organic matter (2.1%), low phosphorus (51 kg/ha), low potassium (436 kg/ha) and low salt contents (0.34 mS/cm).
Field trial was established on date 06.04.2016 with three replications according to randomized block design. The length of each parcel was five meters, each row was consisting of six rows and interrow space was 20 cm. Seeding rate was 30 kg/ha of seed were planted by hand in accordance with TURAN (2019). 40 kg/ha nitrogen and 100 kg/ ha phosphorus were applied at sowing. Field trial was carried out under irrigtaed conditions. Harvests were conducted during the 10% flowering period. Parcels were harvested three times in 2016, four times in 2017 and four times in 2018. Green herbage yield was calculated and followingly 500 grams of green herbage sample was dried at 70 o C for 48 hours to measure dry yields in accordance with TURAN, (2019). CP (crude protein), ADF (acid detergent insoluble fiber) and NDF (neutral detergent insoluble fiber) ratios were determined by using grinded hay samples by NIRS (Near Infrared Spectroscopy) instrument. Using the obtained values, the digestible dry matter (DDM) and the relative feed value (RFV) were calculated by using given equations 1 and 2 (MORRISON, 2003). DDM = 88.9 -(0.779 x ADF) (Eq.1) RFV = ((DDM x DMI (120 / NDF)) / 1.29 (Eq.2) In these equations, DDM=Digestable Dry Matter; RFV=Relative Feed Value; DMI=Dry Matter Intake. JUMP 0.5 statistical program was used for variance analysis and Tukey test was used for comparisons.

Green herbage and hay yields
Statsitically significant differences (P<0.01) were obtained for green herbage yields and hay yields based on genotypes and trial yerswhich are Ciência Rural, v.50, n.11, 2020. Cacan et al. are given in table 2. The majority of the populations yielded higher in terms of green herbage compared to cultivars (Basbag, Bilensoy 80, Savas and Gea). 2016, 2018 and three years average green herbage yields of average of all genotypes were 51.58, 67.91, 96.39 and 71.96 t/ha, respectively. The highest green herbage average yield (96.39 t/ha) was obtained in the third year of the study. 18 of 27 genotypes were yielded high and positioned in the same statistical group according to three years average green herbage yields. Instead, green herbage yields were highest at Mutluca population in the first year (65.41 t/ha), at Garip population in the second year (81.39 t/ha), at Bagliisa population (109.07 t/ha) in the third year, and at Mutluca population for the average of three years (84.29 t/ha).
2016, 2017, 2018 and three years average hay yields of average of all genotypes were 16.11, 26.31, 30.16 and 24.19 t/ha, respectively. The highest hay yield average (30.16 t/ha) was obtained in the third year of the study. 14 of 27 genotypes were yielded high and positioned in the same statistical group according to three years average hay yields. Instead, hay yields were highest at Arslanbeyli population in the first year (19.44 t/ha), at Servi population in the second year (30.32 t/ha), at Garip population (39.02 t/ha) in the third year, and at Mutluca population for the average of three years (27.82 t/ha).
As Total forage hay yield were 5.43, 4.45 t/ha in years 2 and 3, respectively. 5 Genotypes produced average values of 6.5-8.5 t/ha forage dry matter yield. GoKALP et al. (2017) reported that they obtained 11.8-13.3 t/ha green herbage yield in Tokat, in Turkey.

Crude protein ratios and crude protein yields
Statsitically significant differences (P<0.01) were obtained for CP rates based on genotypes,trial yersand genotype x year interactions,and for CP yields based on genotypes andtrial yerswhich are are given in table 3. 2016, 2017, 2018 and three years average CP ratios of average of all genotypes were 24.3, 23.2, 21.6 and 23.0%, respectively. The highest and lowest CP average rates of 24.3% and 21.6% were obtained in the first and third year of the study, respectively.20 of 27 genotypes were yielded high and positioned in the same statistical group according to three years average CP ratios. Instead, CP ratios were highest

Months
- 2016, 2017, 2018 and three years average CP yields of average of all genotypes were 3.74, 6.39, 6.50 and 5.54 t/ha, respectively. The highest and lowest CP yield averages were obtained in the third and first years of the study, respectively. 23 of 27 genotypes were yielded high and positioned in the same statistical group according to three years average CP yields. Instead, CP yields were highest at Arslanbeyli and Bagliisa populations in the first year (both 4.42 t/ha), at Saricicek population in the second year (7.24 t/ha), at Servi population (8.07 t/ ha) in the third year, and at Servi population for the average of three years (6.40 t/ha). Under similar arid climatic conditions in Turkey, SENGUL et al.   ADF and NDF ratios Statistically significant differences (P<0.01) were obtained for ADF based on genotypes, trial yers and genotype x year interactions, and for NDF rates based on genotypes and genotype x year interactions which are given in table 4. 2016, 2017, 2018 and three years average ADF ratios of average of all genotypes were 21.8, 23.6, 23.6 and 23.0%, respectively. ADF ratio averages were same and high for second and third year compared to first year of trials. 21 of 27 genotypes were yielded high and positioned in the same statistical group according to three years average ADF ratios. ADF ratios were lowest(indicator of high quality) at Saricicek-3 population in the first year (18.7%), at Gea variety in the second year (18.7%), at Kumgecit population (20.7%) in the third year, and at Kumgecit population for the average of three years (20.8%).
2016, 2017, 2018 and three years average NDF ratios of average of all genotypes were 39.0, 37.9, 38.8 and 38.6%, respectively. 14 of 27 genotypes were ranked low(indicator of high quality) and positioned in the same statistical group according to three years average NDF ratios. NDF ratios were lowest at Bagliisa population in the first year (35.9%), at Bilensoy 80 and Geavarieties in the second year (both 27.5%), at Savas variety (34.8%) in the third year, and at Gea varity for the average of three years (33.3%). Table 3 -Crude protein ratios and crude protein yields of alfalfa genotypes.

DDM and RFV values
It was found that there were statistically significant (P<0.01) differences among genotypes and average of years in terms of digestible dry matter ratios and relative feed values, but there was no statistically significant difference among years in terms of relative feed value. DDM and RFV values of alfalfa genotypes examined in the study are given in table 5 Table 4 -Acid detergent fiber (ADF) and neutral detergent fiber (NDF) ratios of alfalfa genotypes. Ciência Rural, v.50, n.11, 2020. Cacan et al. year, and at Kumgecit population for the average of three years ( Since the relative feed value is calculated from the ADF and NDF ratios, the difference between these two values changes the relative feed value significantly. In general, the alfalfa plant has an excellent RFV quality grade (170-180 RFV) when it contains 21-22% CP, less than 28% ADF and less than 35% NDF (BoMAN, 2017). When these characteristics are considered together, herbage   quality of alfalfa genotypes that we examined in the study are in excellent grade.

CONCLUSION
The natural variation in wild germplasm is very wide and could be detected and transferred in breeding programs and this variation has been reported earlier in Turkish alfalfa germplasm (SAKIRoGLU et al. 2011). The region that genotypes collected in our study is in the Fertile Crescent and also in the intersection of Vavilov's two centers of origins, Transcaucasia and Mediterranean. The region has been noted to be one of the centers of diversity for alfalfa (SMALL, 2011;SAKIRoGLU & BRUMMER 2013). Germplasm collected showed a desirable charecters for improving yield and quality of alfalfa. Our results indicated that the collection and utilization of wild alfalfa germplasm is crucial for increasing its adaptation. In terms of both green herbage and hay yield, Servi, Mutluca, Garip,Sarıçiçek-1, Bagliisa village populations were found superior to registered varieties. When we compare the quality results, Kumgecit, Küçüktekören and Şenköy-1 populations stands out in terms of crude protein ratio. Highest crude protein yield was obtained from Servi population. In terms of relative feed value, it was observed that the varieties were better qualfied than the populations. It is concluded that the populations of Servi, Mutluca, Garip, Sarıçiçek-1 and Bagliisa can be used in breeding studies especially to obtain genetic progress in yield of alfalfa for feed and bioenergy production. For the genetic improvement of current genotypes with higher crude protein ratio, Kumgecit, Küçüktekören and Şenköy-1 populations can be used as a source. Our policy towards the use of cookies All Clarivate websites use cookies to improve your online experience. They were placed on your computer when you launched this website. You can change your cookie settings through your browser.