PDF version
for printing

Impact
Statement

© 2005 Plant Management Network.
Accepted for publication 16 March 2005. Published 5 April 2005.

Cutting Alfalfa in Late Autumn Increases Annual Yield, Doesn’t Hurt Stands, But is Unlikely to Increase Profit

Robert L. Kallenbach, Associate Professor, C. Jerry Nelson, Professor, and John H. Coutts, Research Associate, Division of Plant Sciences, University of Missouri, Columbia 65211; and Matt D. Massie, Senior Research Specialist, University of Missouri Southwest Research and Education Center, Mt. Vernon 65712

Corresponding author: Robert L. Kallenbach. kallenbachr@missouri.edu

Kallenbach, R. L., Nelson, C. J., Coutts, J. H., and Massie, M. D. 2005. Cutting alfalfa in late autumn increases annual yield, doesn’t hurt stands, but is unlikely to increase profit. Online. Forage and Grazinglands doi:10.1094/FG-2005-0405-01-RS.

Abstract

Alfalfa (Medicago sativa L.) growers wanting to maximize annual yields are often tempted to take an additional harvest in late autumn. Our overall objective was to determine if harvesting alfalfa an “extra” time in late autumn impacts long-term yield and stand persistence. Averaged over five years, the net gain in yield for taking an additional harvest in late autumn was only 0.11, 0.25, and 0.40 tons/acre for alfalfa harvested every 28, 35, and 42 days during the growing season, respectively. Stand persistence was equal for all treatments. An economic analysis shows that the low net yields from a late-autumn harvest usually do not justify the added expense.

Introduction

Alfalfa growers wanting to maximize annual yields are often tempted to harvest during or just after the “critical late-autumn period” (Fig. 1). While the impact of harvesting during or just after this period has been widely studied, recommendations vary by region (5,10,20). Recommendations for the northern U.S. are that alfalfa can be harvested safely in late autumn only if: soil fertility is high (15,20), a disease-resistant cultivar is used (8,15,20), snow cover is adequate (4,15), a late-summer rest period is observed (16,20), and the field is not harvested frequently during the growing season (15,16). If these stipulations cannot be met, a late-autumn harvest is generally not recommended in the northern U.S.

A  B Fig. 1. Alfalfa growers wanting to maximize annual yields are often tempted to harvest alfalfa in late autumn just after the plants go dormant. If high-quality forage is to be obtained, timing of this harvest is critical because nutritive value drops rapidly after plants become dormant. Recommendations about this practice vary by region, with the practice commonly recommended in the southern U.S. and less often in the northern U.S. Figure 1A shows an alfalfa field being baled in late autumn and Figure 1B is a field in early winter that was not harvested in late autumn.

In contrast, research conducted in the southern U.S. shows different results. In Alabama (13), Tennessee (14), Georgia (2), and Oklahoma, (5,18), harvesting alfalfa in late autumn had little effect on stand persistence or yields in following years.

The effects of taking an extra harvest in late autumn are less clear in the transition zone between the southern and northern U.S. Collins and Taylor in Kentucky (3) and Horrocks and Zaifnejad (10) in Utah demonstrated that alfalfa harvested in late August or early September could be safely harvested again in early November once active growth had stopped. However, if the additional harvest was taken in September or October instead of November, yield the following spring was reduced. Horrocks and Zaifnejad (10) also found that older stands of alfalfa were more susceptible to winter-kill when harvested in late autumn than were younger stands. Based on this work, current recommendations in the lower Midwest are that if an additional late-autumn harvest is desired, it be taken in early November. However, to our knowledge there is no research determining the cumulative effects of the practice over several years on the same stand.

Unlike regions where alfalfa is used in a 2- or 3-year rotation with other crops, producers in the lower Midwest expect stands to last 5 years or more. Despite research on the effects of harvesting alfalfa in late autumn, most studies applicable to the lower Midwest have been conducted for 2 years or less on the same field, and therefore do not fully address the long-term and/or cumulative effects of a late-autumn harvest.

Our hypothesis was that in the lower Midwest, harvesting alfalfa an “extra” time in late autumn (early November) would increase annual yields, but as in Kentucky, reduce stand persistence. Additionally, we hypothesized that an “extra” harvest would reduce stands (stem density) most when alfalfa was stressed from being harvested frequently during the previous growing season as seen in the northern U.S. Our primary objective was to determine if harvesting alfalfa an “extra” time in late autumn impacts long-term yield and stand persistence. A secondary objective was to determine if harvest frequency during the growing season alters the response of alfalfa to an extra harvest in late autumn.

Field Experiments on Harvesting Alfalfa in Late Autumn

‘Pioneer 5373’ alfalfa, a multi-pest-resistant hay type, was seeded on 1 April 1994, at the Southwest Missouri Research Center near Mt. Vernon, MO. Soil type was a Huntington silt loam (fine-silty, mixed, mesic Fluventic Hapludoll). Prior to planting, the plot area was sprayed with EPTC (S-ethyl dipropylthiocarbamate) at 3.5 lb/acre a.i. to control weeds during establishment. Seeds were inoculated with Sinorhizobium meliloti, and then seeded at a rate of 15 lb/acre. After seeding, the soil was firmed with a cultipacker. Plots were harvested three times during the seeding year, the last harvest being on 15 September 1994.

Annually throughout the experiment, soil samples were analyzed by the University of Missouri Soil Testing Lab to monitor soil fertility. Based on the results from these samples, lime, phosphorus, potassium, and boron were applied at levels recommended for alfalfa and are outlined specifically by Kallenbach et al. (11) in a related manuscript. Weeds and insects were controlled as described by Kallenbach et al. (11).

There were six treatments consisting of three harvest frequencies during the growing season (28, 35, or 42 days between harvests) as main plots and two late-autumn management regimes (either not harvested or harvested an extra time in late autumn) as sub-plots. This design gave the greatest precision to the late-autumn harvest treatments (19). The dates of initial and subsequent harvests for each harvest frequency were staggered so that the final harvest for each treatment was on 15 September (± 2 days) each year. A further description of harvest dates prior to 15 September are described by Kallenbach et al. (11). September 15 is the currently recommended date for a final harvest during the growing season in Missouri (9). By harvesting all plots on 15 September, there were an equal number of growing days in late autumn to measure the residual effects of previous harvest frequencies. Late-autumn harvests were taken in late October or early November each year (Table 1). Late-autumn harvest treatments began in 1995 and continued through 1999.

Table 1. Date of late-autumn harvest for all treatments, date of first killing frost (23°F), growing degree days (GDD, 40°F basis) from 15 September to the late-autumn harvest date, and GDD accumulated after the late-autumn harvest date until the first killing frost.

^y First date in late-autumn when temperature dropped below 23°F.

^z Average total accumulated precipitation for September and October at Mt. Vernon, MO is 8.6 inches.

Individual sub-plots were 6 ft × 25 ft. Yields were measured by cutting a 2.7-ft-×-20-ft strip from the center of each plot with a flail-type harvester set to leave a 2.5-inch stubble. Fresh mass of the harvested forage was recorded. A sub-sample (approximately 0.75 lb of fresh mass) from each strip was weighed fresh and again after drying for 96 h or more in a forced-air oven at 122°F to determine dry matter. After drying, samples were ground in a cyclone mill (UDY Corp., Ft. Collins, CO) to pass a 0.04-inch (1-mm) screen and used to evaluate forage nutritive value.

Forage Nutritive Value Analysis

Crude protein and neutral detergent fiber were determined for samples collected in the late autumn of 1995, 1997, 1998, and 1999 using near infrared reflectance (NIR) spectroscopy. Forage samples from 1996 were accidentally discarded. The NIR spectrophotometer was a Pacific Scientific 6250 scanning monochromator (NIRSystems, Silver Spring, MD) operating with software developed by Infrasoft International (Port Matilda, PA). The spectrophotometer was calibrated for crude protein and neutral detergent fiber by regressing chemically-derived data against spectral data using modified partial least squares regression (17).

Crude protein for calibration samples was determined by measuring forage N content using a Leco FP-480 Nitrogen Analyzer (Leco Corp., St. Joseph, MI), and crude protein was calculated as percentage N × 6.25. Neutral detergent fiber for calibration samples was determined using the methods described by Goering and Van Soest (7).

Stem Density Measurements

Stem density was determined by counting the number of stems in four randomly-placed, 1-ft² quadrats within each plot. Stem density was measured in June and in December of the establishment year (1994). Thereafter, stem density was measured immediately after harvest in May and September each year.

Economic Analysis

Grower returns were analyzed using partial budgeting. Partial net returns for a late-autumn harvest were calculated as additional gross returns minus additional costs. Gross returns were calculated as additional yield per acre (or “net yield”) times price per ton. We used a price of $90/ton, a long-term average for “dairy quality” hay in Missouri (1).

To assess the costs for a late-autumn harvest, our analysis included additional harvest costs and fertilizer needed to replace nutrients removed. Harvest and fertilizer costs were obtained from a recent economic analysis of alfalfa production from the University of Minnesota (12) and modified for conditions in the lower Midwest (1). In this analysis, mowing and conditioning cost $5.53/acre, tedding $4.27/acre, raking $5.34/acre, and baling $7.64/acre. For mowing and conditioning, tedding, raking, and baling it was assumed that each operation would be done once for the late-autumn harvest and the cost per acre is independent of yield. Hauling costs were $16.67/ton. Costs include labor, fuel, and repairs but do not include any additional overhead or fixed costs. This approach was used as harvesting alfalfa an “additional” time in late autumn would likely use existing equipment and we wanted to capture only the additional costs in our partial budget. Fertilizer costs were $0.22/lb for P₂O₅, and $0.15/lb for K₂O. For each ton of alfalfa harvested as hay, it was assumed that 15 lb of P₂O₅, and 60 lb of K₂O would need to be applied to replace the nutrients removed (1). No additional costs were assigned for weed or insect control.

Experimental Design

The experimental design was a randomized complete block in a split-plot arrangement of treatments with four replications. Analysis of variance was conducted on harvest frequencies (main plots), late-autumn harvest managements (sub-plots), and years (sub-sub-plots) and all possible interactions using the model outlined by Steel and Torrie (19). The Proc GLM function of SAS (version 8) was used for statistical analyses (SAS Institute, Cary, NC). Main effects and all interactions were considered significant when P < 0.05. When the F test was significant (P < 0.05), means were separated using Fisher’s protected LSD (alpha = 0.05) (19).

Late-Autumn Forage Yields and its Impact on the Following Year

Harvesting alfalfa less frequently during the growing season always led to greater yields in late autumn (Fig. 2, top panel). Averaged over all years, plots harvested every 28 days during the growing season yielded only 0.28 tons/acre in late autumn, compared to 0.44 and 0.56 tons/acre when harvested every 35 or 42 days, respectively. The linear relationship (r² = 0.99, b = 0.020, P = 0.06) between previous harvest frequency and late-autumn yield demonstrates the residual effect of in-season harvest management. Yields in late autumn were positively associated with accumulated rainfall during September and October (r² = 0.86, b = 0.042, P = 0.02 over all harvest frequencies), but not with the number of growing degree days (40°F basis) after 15 September (r² = 0.03; NS) (Table 1).

We also calculated the impact of a late-autumn harvest on forage yield the following year. Yield change at the first harvest the following spring of plots harvested in late autumn when compared to plots not harvested in late autumn is shown in Fig. 2 (middle panel). Yield loss due to a late-autumn harvest was minor (≤ 0.15 tons/acre) in the spring of 1996, 1997, or 2000. In the spring of 1998 and 1999, however, yield of plots harvested in late autumn showed reductions of 0.25 to 0.61 tons/acre. Harvest frequency during the growing season had no effect on the relative yield reduction the following spring, except for the spring of 1999. While yield in late autumn is consistently influenced by harvest frequency, the yield reduction the following spring is largely independent of harvest frequency. Research conducted at a variety of locations in the U.S. and Canada show that weather conditions during winter and the following spring play a large role in determining the impact of a late-autumn harvest (4,13,15). Within a harvest frequency, there was no effect of a late-autumn harvest on the second or subsequent harvests each year and no interactions between years. As reported by Kallenbach et al. (11) in a related study, averaged over years, yield for the rest of the season (first and late-autumn harvests excluded) for the 28-, 35-, and 42-day harvest frequencies was 2.62, 3.37, and 3.38 tons/acre, respectively, when alfalfa was not harvested in late autumn versus 2.49, 3.35, and 3.33 tons/acre for alfalfa harvested in late autumn.

Net autumn yield (autumn yield plus or minus the change in yield at the first harvest in spring) varied substantially by year, with the largest net yields recorded in 1996-1997 (Fig. 2, bottom panel). Net yield for 1996-1997 was 0.28, 0.61, and 0.94 tons/acre for alfalfa harvested every 28, 35, or 42 days during the growing season, respectively. Averaged over all five years, net yields were 0.11, 0.24, and 0.40 tons/acre for harvest frequencies of 28, 35, and 42 days, respectively. The low net yields for all treatments and the likelihood that variation between years for net yield depends largely on conditions during the subsequent winter make it difficult to predict when a late-autumn harvest might be most productive.

Forage Nutritive Value

There was no interaction between years and previous harvest frequency for crude protein or neutral detergent fiber from alfalfa harvested in late autumn, so the means over all years are presented in Fig 3. Nutritive value was generally high for all treatments but was greatest (low neutral detergent fiber and high crude protein) from plots harvested 28 days throughout the growing season and lowest when harvested every 42 days. However, the differences between harvest frequencies were small and it is doubtful that they represent any meaningful biological differences or economic opportunities for producers. Sheaffer et al. (16) also reported the high nutritive value of alfalfa harvested in late autumn, noting that it provided some of the best quality forage of the entire growing season.

Except for the spring of 1997, forage nutritive value at the first harvest the following spring was generally equal or greater for plots harvested in late autumn than plots not harvested in late autumn (Fig. 4). There was no interaction between harvest frequency and late-autumn harvesting for nutritive value at the first harvest in spring so the data were combined over harvest frequencies. We observed dead stems from the previous autumn’s growth in the plots that were not harvested in late autumn and this may have reduced forage nutritive value at the first harvest the following spring in those treatments. Another possibility is that since plots not harvested in late autumn usually yielded more the subsequent spring (Fig. 2), the increased vigor in these treatments caused plants to mature more rapidly in spring when compared to plots harvested in late autumn. The higher nutritive value of plots harvested in late autumn could have resulted from both not having residue left from the previous autumn as well as less mature plants due to slower regrowth in spring. After the first harvest in spring, there were no differences in nutritive value between late-autumn harvest treatments (data not shown).

Effect of Late-Autumn Harvesting on Stem Density

Stem density was unaffected by harvesting in late autumn or by harvest frequency. Over the life of the study, the number of stems per square foot declined from 61 in 1995 to 36 by 2000. This shows that alfalfa can be harvested in late autumn without sacrificing stem density in the lower Midwest. This is similar to reports from Michigan (20), Alabama (13), Georgia (2), and Oklahoma (18) where harvesting alfalfa in late autumn had little impact on stand survival. Our data from a different environment corroborate these results and indicate that yield, not stem density, is the major factor affected the year following a late-autumn harvest.

Economic Analysis

Averaged over all years, harvesting alfalfa in late autumn is more likely to decrease than increase net returns, but projected return depends strongly on harvest frequency during the growing season (Table 2). For alfalfa harvested every 28 or 35 days during the growing season, an additional harvest in late autumn decreases net returns by $16.07/acre and $7.52/acre, respectively. This is primarily due to low net yields and harvest costs that are insensitive to yield (for example, the cost to mow and condition is the same regardless of yield). Net returns are increased by $1.63/acre if a 42-day harvest frequency is used. Cash hay producers often like to harvest in autumn because they can sell high quality hay at a premium; however, the break-even prices demonstrate that fall harvesting is unlikely to be profitable when harvest intervals during the growing season are less than 35 days. The overall low net yields from a late-autumn harvest are not likely to substantially increase profits compared to other management decisions such as pest management, cultivar selection, and harvest frequency (5,11,15,20).

Table 2. Partial budget showing additional gross returns, costs, net returns and break-even cost/ton for alfalfa harvested an additional time in late autumn.

Additionally, there are several factors that make a late-autumn harvest risky in the lower Midwest, notably, variable net yields from year to year (Fig. 2) which lead to wide annual variations in profitability. The coefficients of variation for net returns were high between years: 42, 172, and 1117% for the 28-, 35-, and 42-day harvest frequencies, respectively. Also, the unfavorable weather conditions for haymaking in late autumn increase risk. Other low-cost harvesting methods, such as grazing, might offer more consistent or greater economic returns (6,8). Grazing in late autumn would have the advantage of lower harvest costs, and less sensitivity to poor weather conditions. Other research shows grazing in late autumn reduces alfalfa weevil populations the following spring (6) and is less likely to reduce yields in the following year when compared to mechanical harvest (8).

Conclusions and Recommendations

• Autumn yields are greatest when the stand is harvested infrequently during the growing season and rainfall is adequate in September and October.

• Forage nutritive value of alfalfa in late autumn is typically high. Although alfalfa harvested frequently during the growing season is slightly higher in nutritive value than alfalfa harvested infrequently, the difference is of little biological or economic importance.

• Stem density is not reduced by harvesting in late autumn.

• In 2 of 5 years, taking an extra harvest in late autumn reduced forage yields the following spring, which lowered net yields. Further, the residual effect of a late-autumn harvest is unpredictable prior to making a harvest decision. This uncertainty makes it difficult to advise producers as to which years this practice might be most profitable.

• If alfalfa is harvested every 42 days during the growing season, then a late-autumn harvest can increase profits, but the net returns are likely to be variable and small. A majority of producers in the lower Midwest use a 28- to 35-day harvest frequency during the growing season; when harvesting this frequently during the growing season, net returns would likely be reduced by taking a late-autumn harvest.

Literature Cited

1. Brees, M. 2003. Alfalfa production and alfalfa establishment budgets of southern Missouri. MU Extension Publ. FBM3101. Univ. Missouri, Columbia, MO.

2. Brown, L. G., Hoveland, C. S., and Karnok, K. J. 1990. Harvest management effects on alfalfa yield and root carbohydrates in three Georgia environments. Agron. J. 82:267-273.

3. Collins, M., and Taylor, T. H. 1980. Yield and quality of alfalfa harvested during autumn and winter and harvest effects on the spring crop. Agron. J. 72:839-844.

4. Dhont, C., Castonguay, Y., Nadeau, P., Belanger, G., Drapeau, R., and Chalifour, F. P. 2004. Untimely fall harvest affects dry matter yield and root organic reserves in field grown alfalfa. Crop Sci. 44:144-157.

5. Dowdy, A. K., Berberet, R. C., Stritzke, J. F., Caddel, J. L., and McNew, R. W. 1993. Interaction of alfalfa weevil (Coleoptera: Curculionidae), weeds, and fall harvest options as determinants of alfalfa productivity. J. Econ. Entomol. 86:1241-1249.

6. Dowdy, A. K., Berberet, R. C., Stritzke, J. F., Caddel, J. L., and McNew, R. W. 1992. Late fall harvest, winter grazing, and weed control for reduction of alfalfa weevil (Coleoptera: Curculionidae) populations. J. Econ. Entomol. 85:1946-1953.

7. Goering, H. K., Van Soest, P. J. 1970. Forage fiber analyses (apparatus, reagents, procedures, and some applications). Agric. Handbook 379. ARS, USDA, Washington, D.C.

8. Gray, F. A., and Koch, D. W. 2004. Influence of late season harvesting, fall grazing, and fungicide treatment on Verticillium wilt incidence, plant density, and forage yield of alfalfa. Plant Dis. 88:811-816.

9. Henning, J. C., and Nelson, C. J. 1988. Alfalfa. MU Extension Publ. G4550. Univ. Missouri, Columbia, MO.

10. Horrocks, R. D., and Zaifnejad, M. 1997. Late-season management of alfalfa in irrigated valleys of the intermountain west. J. Prod. Agric. 10:96-101.

11. Kallenbach, R. L., Nelson, C. J., and Coutts, J. H. 2002. Yield, quality, and persistence of grazing- and hay-type alfalfa under three harvest frequencies. Agron. J. 94:1094-1103.

12. Lazarus, W. F., and Selley, R. A. 2003. Farm machinery economic cost estimates for 2004. Univ. Minn. Exten. Publ. FO-6696. Univ. Minn., St. Paul, MN.

13. Mays, D. A., and Evans, E. M. 1973. Autumn cutting effects on alfalfa yield and persistence in Alabama. Agron. J. 65:290-292.

14. Reynolds, J. H. 1971. Carbohydrate trends in alfalfa (Medicago sativa L.) roots under several forage harvest schedules. Crop Sci. 11:103-106.

15. Sheaffer, C. C., and Marten, G. C. 1990. Alfalfa cutting frequency and date of fall cutting. J. Prod. Agric. 3:486-491.

16. Sheaffer, C. C., Wiersma, J. V., Warnes, D. D., Rabas, D. L., Lueschen, W. E., and Ford, J. H. 1986. Fall harvesting and alfalfa yield, persistence, and quality. Can. J. Plant Sci. 66:329-338.

17. Shenk, J. S., and Westerhaus, M. O. 1991. Population structuring of near infrared spectra and modified partial least squares regression. Crop Sci. 31:1548-1555.

18. Sholar, J. R., Stritzke, J. F., Caddel, J. L., and Berberet, R. C. 1988. Response of four alfalfa cultivars to fall harvesting in the southern plains. J. Prod. Agric. 1:226-270.

19. Steel, R. G. D., and Torrie, J. H. 1980. Principles and procedures of statistics: A biometrical approach. 2nd ed. McGraw-Hill, New York.

20. Tesar, M. B., and Yager, J. L. 1985. Fall cutting of alfalfa in the North Central USA. Agron. J. 77:774-778.