Using Stem Count as a Decision Aid in Turning Over Alfalfa Stands

Dan Undersander, Department of Agronomy, University of Wisconsin, Madison 53706; and Dennis Cosgrove, Department of Plant and Earth Science, University of Wisconsin, River Falls 54022

Abstract

Alfalfa stem density can be an indicator of when alfalfa stands no longer have potential for high yield. Studies were conducted at University of Wisconsin Arlington Research Station and River Falls Farm where areas of differing plant density were selected within 1-, 2- and 3-year-old fields of differing varieties each year, for three years. Stem density and yield were determined on the same plot areas for each cutting and plant density was determined at the end of the season. Plant density and stems per plant had little relationship to yield. Stem density below 40 stems/ft² had reduced yield potential.

Introduction

The decision to turn over an alfalfa stand, based on economics, is largely when alfalfa yield begins to decline as stands thin. An additional factor is that weeds begin to invade in thinning stands. The first effect of stand thinning is often not total yield reduction because loss of alfalfa tonnage is replaced by weeds, but rather a loss in forage quality since most weeds have lower quality than alfalfa when it is harvested (3,4).

Since alfalfa yield potential can change significantly from year to year, a previous year’s yield is of little value in estimating the next year’s yield. Thus, until recently, most managers have considered that yield is a function of stand density and used plant density counts to estimate the alfalfa stand’s productive ability.

Density of stand required for maximum yield varies depending on geographic area and climate. The plant stand density producing maximum yield the year after reseeding has been reported to be 13 plants/ft² in California (8) and Michigan (9), 23 in Ohio (14), and 26 in Illinois (6) and 17 plants/ft² in Indianna (15).

However, as alfalfa stands age, individual plants grow larger and fewer plants per square foot are necessary for maximum yield (5,10,13). However, neither relationship is well defined and variability, depending on plant size, leads to poor relationships between plant density and yield. Furthermore, visual plant counts estimates may not be accurate (1) due to inability to identify from above ground intertwined plants crowns and split crowns.

Yield can be thought of as a function of plant components that includes plant number per unit area, stem number per plant, and stem weight (5). However, the above data and other research (15,16) have shown that individual components may have low correlations with plant yield. The objective of this study was to determine if plant stand components could be used to develop a response curve, indicating when alfalfa stand density was limiting yield.

Measuring Yield-Stem Denisty Correspondence

These studies were conducted on production alfalfa hay fields at the University of Wisconsin River Falls Farm (44°51’N, 92°37’W) on a Nickin silt loam (fine-loamy over sandy or sandy-skeletal, mixed, superactive, frigid Typic Argiudoll) and the University of Wisconsin Arlington Research Station (43°18’N, 89°21’W) on a Plano silt loam (fine-silty, mixed, mesic. Typic Argiudoll). Each year for three years, first, second, and third production-year stands were selected at each site (total of 9 fields per year per site). The first production year is defined as the year after seeding. Fields were soil tested and selected where soil fertility was in the optimum to high range. An effort was made to have fields representing different alfalfa cultivars and 23 commercial cultivars with a range in fall dormancy of 2 to 4, recommended during the years of the study, were involved. Nine sites (10.8 ft²) were selected within each field having a visual range in plant density and were permanently marked. Herbicide was applied if necessary to assure weed-free stands and fields were scouted and sprayed if necessary to minimize insect losses. Every effort was made to ensure optimum forage growth. During the three years of the study there was no significant water stress at any site. Sites were harvested immediately prior to harvest of the entire field (generally at first bloom). Harvestable stems (stems above cutterbar height) were counted and then dried at 140°F for yield determination. At the end of the season, in 5 of 6 site-years, plants were dug and individual plants counted. Plant count data was not collected at River Falls in 1992 due to early ground freeze. Total season yield rather than individual cutting yields were used as the dependent variable because the fit was better than if yield from individual cuttings were used (data not presented). Alfalfa stands in northern states tend not to change in plant count during the growing season as they do in states to the south. In northernmost states plants become diseased over summer and die over winter. Therefore total season yield was more related to plant components than yield of the last cutting taken immediately before destructive plant counts were taken. Total season yield is also the variable that farmers are interested in and is more related to profitability than individual cutting yield. Data was analyzed with PROC MIXED (SAS Institute Inc., Cary, NC); a type III analysis was used to adjust for unequal sample sizes. Locations were considered a fixed effect and other factors (years, varieties, stand age) random.

Alfalfa Stem Density as Indicator of Yield Potential

Since trend over time was not an objective of this study, the data does not follow a given stand over years as different fields were selected for each year of the study. Also, since recommendations for turning over stands must be able to be made across cultivars, an effort was made to include many different alfalfa cultivars representing the major commercial breeding programs. However, multiple fields of several of the varieties were sampled during this study. In no case, were any significant varietal differences in response of plant components noted (data not presented). Thus while plant component response may vary among a wider range of germplasms, differences were not observed over the range of commercial cultivars with 2 to 4 fall dormancy tested in this study.

In this study plant density generally declined with age of stand from 9.7 plants/ft² to 6.6 and 6.1 plants/ft² in the second and third production years of stand (Table 1). This data agrees with the general trend of plant density decline as a stand ages (5,10,13). However, plant density numbers in this study (Year 1) are much less than the 14 to 26 plants/ft² reported as needed the year after seeding for high yield by Tesar and Marble (10).

Table 1. Effect of stand age on plant components. Values are means of all plots harvested in stand age.

Stand age (year)	Plants/ft²	Stems/plant	Stems/ft²	Yield (tons/acre)
1	9.7a*	4.8b	44a	4.24a
2	6.6b	5.5ab	34b	3.53b
3	6.1b	5.6a	28c	3.30b

* Values within columns followed by different upper case letters and within rows followed by different lower case letters differ according to least significant difference multiple range separations (P < 0.05).

In this study plants per square foot and stems per square foot were generally highest in the first production year (Figs. 1a, 1b, 2a, and 2b). Plant and stem density was variable and overlapping in second and third production years.


Fig. 1a. Arlington plant counts and total season yield by stand age.		Fig. 1b. River Falls plant counts and total season yield by stand age.


Fig. 2a. Arlington season mean stem counts and total season yield by stand age.		Fig. 2b. River Falls season mean stem counts and total season yield by stand age.

Figures 3a and 3b show why plant density is not a good indicator of when to turn over fields of alfalfa. While the relationship is significant (P > 0.01), the coefficients of determination were only 0.255 and 0.195, respectively for the two locations. The highest yielding plots generally had around 10 plants/ft². However, high yields were obtained with much lesser plant populations. More importantly, there was tremendous variation in yield among the stands with 10 plants/ft², ranging from 2 to 7 tons/acre at each location. Thus the common recommendation of 6 plants/ft² being adequate for high yield may or may not be correct depending on the size of the plants.


Fig. 3a. Arlington plant counts vs total season yield.		Fig. 3b. River Falls plant counts and total season yield.

The number of stems per plant increased with stand age from 4.8 to 5.5 and 5.6 (Table 1) in agreement with what was reported by Fick et al. (5). Presumably crowns were increasing in size as the plant aged. Stem number per plant averaged only 5.3 in this study which is slightly lower than reported by Volenec et al. (15) at similar plant densities though, some relatively high yields were obtained with the stand densities of this study (Figs. 3 to 5). In this study there was no relationship between stems per plant and yield (Figs. 4a and 4b). Fick et al. (5) reported that the stem number was reduced by moisture stress (too much or too little), lack of sunlight and other stresses. Stem initiation and development appears to be extremely environmentally dependent and therefore would not be a good indicator of stand yield potential.


Fig. 4a. Arlington stems per plant and total season yield.		Fig. 4b. River Falls stems per plant and total season yield.


Fig. 5a. Arlington season mean stem counts and total season yield.		Fig. 5b. River Falls season mean stem counts and total season yield.

The relationship of stems per ft² to yield is presented in Figures 5a and 5b. Of the parameters measured in this study, stem density had the best relationship to yield with coefficients of determinations of 0.569 and 0.491 for Arlington and River Falls, respectively. Berg et al. (1) reported that stem density had no relationship to alfalfa yield and that stem mass was the primary plant component related to yield. We agree that when plant stresses, such as nutrient deficiency, drought, or insect pressure, occur stem growth is the first to respond. This would indicate that stem mass would be the best indicator of yield, especially within a cutting as the authors indicated. Our purpose, however, was not to estimate yield but to determine when the stand would be limiting yield with other yield affecting factors minimized. It appears that use of stem density is well suited for this purpose.

Stem density was the best indicator of overall stand vigor and yield potential among the plant component variables examined. Breakpoint analysis indicated that stands with less than 40 stems/ft² had insufficient stem density for highest yields in this study (Figs. 5a and 5b). In two of the three years, the response was linear up to 55 stems/ft². However the higher stem density was generally in first production year stands when turn over is generally not a consideration (Figs. 2a and 2b). Actual yield will depend on stem mass since growth of stems is more immediately sensitive to environmental variables such as drought, nutrient deficiency, and other stresses.

Future yield estimates are the crucial decision when deciding whether or not to keep an alfalfa stand as it is being evaluated in the fall or spring. Plant density estimates have been useful for this purpose but are not well correlated with future yield potential because plant number changes as stands age and crown size changes variably. Low plant densities can produce high yields if plants are large and healthy producing many stems as shown in this study. Similar plant densities that are less vigorous will have low yields. This variability in plant size and stem production accounts for the wide range of yields observed at similar plant densities.

Stands with lower yield potential should be rotated to another crop since a large portion of the cost of forage production is fixed and high yields generally result in the least production cost per ton of forage. Keeping stands with low forage yield potential raises the cost of forage production per ton produced. Turning over low yield potential stands will also give the grower legume credits for the next crop and rotational benefits of higher yield for the next crop (7,11,12).

Stem density is an easy parameter to estimate in the field. Accurate plant counts are difficult due to inability to discern crowns from single versus multiple plants in older fields (1). Stem density can easily be counted immediately after harvest by counting cut ends or by counting first season growth when plants are 6 to 8 inches tall. With very little experience an individual can train themselves to do visual estimates of stem density. This allows entire fields to be quickly surveyed and to determine the percentage or portions of the field in adequate or in stand limiting ranges. Then the grower can make the decision on what to do with the field based on the proportion of the field with greater than 40 stems/ft².

Literature Cited

1. Berg, W. K., Cunningham, S. M., Brouder, S. M., Joern, B. C., Johnson, K. D., Santini, J., and Volenec, J. J. 2005. Influence of phosphorus and potassium on alfalfa yield and yield components. Crop Sci. 45:297-304.

2. Bundy, L. G., Kelling, K. A., and Good, L. W. 1997. Using legumes as a nitrogen source. Univ. of Wis. Ext. Publ. A3517. Madison, Wis.

3. Cosgrove, D. R., and Barrett, M. 1987. Effects of weed control in established alfalfa (Medicago sativa) on forage yield and quality. Weed Sci. 35:564-567.

4. Doll, J. D. 1986. Do weeds affect forage quality? Pages 161-170 in: Proc. of 16th Nat. Alfalfa Symp., Fort Wayne, IN.

5. Fick, G. W., Holt, D. A., and Lugg, D. G. 1988. Environmental physiology and crop growth. Pages 169-170 in: Alfalfa and Alfalfa Improvement. Agron. Monogr. 29. A. A. Hanson, D. K. Barnes, R. R. Hill Jr., eds. ASA, CSSA SSSA. Madison, WI.

6. Jackobs, J. A., and Miller, D. A. 1979. Varying seeding rates of alfalfa. In 1979 Agron. Abstr. ASA, Madison, WI.

7. Kelner, D. J., Vessey, J. K., and Entz, M. H. 1997. The nitrogen dynamics of 1-, 2- and 3-year stands of alfalfa in a cropping system. Agric. Ecosyst Environ. 64:1-10.

8. Marble, V. L., and Peterson, B. 1981. Planting dates and seeding rates for central California. Pages 22-26 in: Proc. of the 11th Calif. Alfalfa Symp., Fresno, CA

9. Tesar, M. B. 1977. Productivity and longevity of alfalfa as affected by date and rate of seeding, variety and annual topdressing. Pages 12-14 in: Proc. of the 15th Central Alfalfa Improvement Conf., Ames, IA.

10. Tesar, M. B., and Marble, V. L. 1988. Alfalfa establishment, in alfalfa and alfalfa improvement. Pages 303-304 in: Agronomy Monograph 29. A. A. Hanson, D. K. Barnes, and R. R. Hill Jr., eds. ASA, CSSA SSSA, Madison, WI.

11. Smith, M. A., Carter, P. R., Imholte, A. A. 1992. Conventional vs. no-till corn following alfalfa/grass: Timing of vegetation kill. Agron. J. 84:780-786.

12. Smith, N. A., and Carter, P. R. 1993. No-till systems for corn following hay or pasture. J. Prod. Agric. 6:46-52.

13. Sund, J. M., and Barrington, G. P. 1976. Alfalfa seeding rates: Their influence on dry matter yield, stand density and survival, root size, and forge quality. Univ. of Wisc., Coll. of Agric. and Life Sci., Res. Bull. R2786. Madison, WI.

14. Van Kuren, R. W. 1973. Alfalfa establishment and seeding rate studies. Ohio Rep. 58:52-54.

15. Volenec, J. J., Cherney, J. H., and Johnson, K. D. 1987. Yield components, plant morphology, and forage quality of alfalfa as influenced by plant population. Crop Sci. 27:321-326.

16. Volenec, J. J., and Cherney, J. H. 1990. Yield Components, morphology and forge quality of multifoliolate alfalfa phenotypes. Crop Sci. 30:1234-1238.