Troubleshooting concerns associated with direct long cut vacuum grass silage

Final Report for ONE06-059

Project Type: Partnership
Funds awarded in 2006: $7,572.00
Projected End Date: 12/31/2007
Region: Northeast
State: Massachusetts
Project Leader:
Sue Ellen Johnson
New England Small Farm Institute
Expand All

Project Information


Direct Long Cut Vacuum Grass Silage (DLCVGS) is a low input method of preserving grass silage. Farmers frequently ask the question of “what is enough vacuum”; or will more vacuuming make a difference. After several years of successful monitored on-farm non-replicated testing, experts voiced concern about possible clostridrial issues because of the relatively high moisture and butyric acid values in the silage products. Clostridia botulinum is a common bacteria found naturally in farm environments. Traditional chopped silage and haylage has been reported to be associated with catastrophic incidents of clostridrial contamination when “wet”. In chopped silage high moisture is usually also associated with higher pH values, high butyric acid and higher ammonia-N values. We proposed to study this issue in DLCVGS a high moisture silage. We also proposed to monitor seepage and exudates from silage piles.

Project Objectives:

To determine if “more vacuum” would alter the quality of the silage or affect the counts of Clostridia/anaerobic bacteria in the silage.

To determine if there was seepage or exudates from DLCVGS piles.


Click linked name(s) to expand
  • Roy Bergeron
  • Arnie Voeringer


Materials and methods:

We followed a technique developed over several years, now in use by a small number of farmers throughout southern New England. In 2006, we created sample piles. We followed the technique as outlined and field-tested. Silage was cut in mid-August (second cut) with a modified older New Holland forage chopper throwing directly into a small wagon. Test piles were 25 x 20. Four trailer loads of forage were formed into each pile with a skid-steer. Piles were immediately covered with 5.5 ml black on white plastic anchored with sand and compost. Then vacuum was applied for each treatment.

The experiment consisted of three replicates (an RCBD with two treatments. Six silage piles were made from a single mixed grass field with minimal percentage of weeds (<6%)and legumes (<5%). After pile building and sealing, one pile in each replicate pair was vacuumed with single shop vac for 20 minutes. The other pile in each pair was vacuumed with two shop vacs for 35 minutes (or until the vacs blew electrical circuits). Two pairs of piles were placed on mowed ground. The third set was placed on asphalt tarmac. Piles were observed daily for damage to plastic, entry of outside air, and the seepage of exudates. In February 2007, piles were opened and 8-10 grab samples were taken and composited from each pile for analysis. Forage samples were sent to a commercial lab (Dairy One). After extensive consultation with several food scientists/microbiologists it was determined to send samples for a general count/evaluation of anaerobic and aerobic bacteria. (The initial lab consulted for this proposal was no longer running clostridial analyses.) The initial plan to analyze for clostridia had been determined to be impractical.

Research results and discussion:

Forage quality changed very little during the ensilage period. Protein percentages increased slightly in several piles. This is consistent with our previous experience. There were no meaningful differences in forage quality between the single vac and double (super) vac piles. Minor differences in silage quality seem to correspond to variability of the initial, fresh forage ensiled, as observed in our previous field tests. Percent moisture in the final product ranged from 72.9 to 81.2.
As has been our previous experience, the silages smelled sweet and pickly, and cattle readily consumed all silage from all the piles.

Volatile fatty acid (VFA) values for the silage were, as has been our previous experience, difficult to interpret using the literature for traditional chopped silages. Percent moisture was higher than generally considered desirable. Percent butyric acid was higher than generally desired (the “goal”), but lower than typical values for high moisture silages. No butyric odors were detected (this has been our experience throughout the series of field tests).

There was minimal (virtually no) mold or other losses in the sample piles. On other tests we have seen small (softball-size) pockets of mold and some at the base of the piles.

Daily observations for one month following the formation of the piles and weekly thereafter did not detect any exudates seeping from the silage piles on the asphalt or on the mown ground. (Again this is consistent with earlier experience). This appears to be an environmentally sensitive technique of silage-making. The one pass system reduces fuel use. We speculate that the lack of exudates and seepage is related to the long cut- crimped, not chopped system of cutting. This prevents loss of nutrients as well as problems of pollution. Both at pile sealing and opening the silage is not sticky or “wet”. Most of the plants’ moisture remains in the plant stems, leaves and cells. The plant structures and individual cells are intact, not chopped, packed or mashed. Vacuuming instead of physical packing also preserves plant structures. We feel these “mechanical facts” fundamentally alter the nutrients available in the pile for fermentation, microbial activity and populations in the pile, and consequently the overall fermentation process.

Our bacterial measurements were unsuccessful. Our initial intent to assay for presence of clostridial bacteria was overambitious for funding and expertise available. We initially were misinformed and underestimated the complexity of the task. (This was communicated to NESARE administration at the time). Adequate “sampling” of the piles would have involved analyzing all the silage in the piles and live assays for botulism toxin, not simply the Clostridial organisms as we had initially thought. The secondary samples sent and analyzed for anaerobic: aerobic analysis were inconclusive due to shipping problems, and the fact that the assay was really not appropriate for a bacterial product like silage.

An alternative, more adequate experiment would have had treatments designed to create favorable conditions for C. botulinum as well as “controls”- our standard practice. This may have required confined facilities. This was not something we were willing to consider.

The entire discourse over botulism episodes related to the intake of spoiled silage by cattle is quite interesting. The incidence is not thoroughly understood. Contamination with C. botulinum does occur – though rarely. Incidents that do occur are catastrophic. Multiple animals die. Theoretically, humans handling the silage might also be affected. The botulism toxin is not necessarily found throughout the silage pile, but only in the area of the pile where the specific environment favors it. Incidents have been traced back to chopped corn and grass silages. No incidents specifically related to DLCVGS have been reported.

There is a theory, discussed widely in the scientific literature, that relates botulism episodes not just to silage moisture but to the presence of animal tissue- dead rodents etc. chopped into the silage at ensiling. Farmers should be cautioned. The episodes have also been related to high percentage levels of butyric acid. We have cautioned farmers not to feed the silage which has the nasty butyric acid odor. Purely anaerobic high moisture conditions have also been suggested as contributing to incidents.

It is possible that handling the silage during the process of feeding out the silage, may reduce the risk because of aeration with tumbling etc.

It does not seem practical to routinely test for C. botulinum (regardless of budget). We have not yet identified another, indicator bacteria for C. botulinum. This would be worth pursuing because of the potential catastrophic consequences of a ”toxic” event.

Research conclusions:

Impacts of results

This was the only fully conventionally replicated study done with DLCVGS. The results parallel our experience in multiple farm tests. We can better inform producers of the risks related to C. botulinum not through our experimental results, but the learnings gleaned through this project. Our final report on DLCVGS (to be posted on the NESFI website) will include this information.
We hope it will attract proper attention of fermentation microbiologists (and we will bring it to their attention), because there seems to be much more to the story of DLCVGS fermentation. Better understanding silage fermentation is important to the future of grass-based enterprises.
Our findings should not discourage producers or advisors from the DLCVGS technique.

Participation Summary

Education & Outreach Activities and Participation Summary

Participation Summary

Education/outreach description:

This project’s findings are being included in a summary report on 5 years experience with DLCVGS to be available via the NESFI website.

An informal field day was held in August when making the silage (4 attendees).

A more formal field day (frigid with chilling winds) was held when the piles were opened and sampled (6 attendees).

Project Outcomes

Project outcomes:

Farmer Adoption

The project findings will better inform farmers using or interested in the DLCVGS. The findings will not be a primary constraint to farmers interested in adopting this technology. Issues related to the logistics of feeding DLCVGS are more of a concern. The interest in this technique seems to be widely dispersed. We get numerous calls from across the country, so the impact will not be particularly regional.

The technique continues to be promulgated in the popular and alternative farm press.
An outstanding issue is that professionals, experts and advisors often disparage the use of the technique based simply on the high moisture and butyric acid levels in the DLCVGS. Their caution makes sense in the context of chopped silages.

Our experience, and the findings of this project strongly suggest the process and products (and potential advantages) are sufficiently different that they merit further investigation before dismissal. It is interesting to note the widespread use of the technique in the British Isles and Europe with minimal to no formal literature on the technique.

Assessment of Project Approach and Areas of Further Study:

Areas needing additional study

Better understanding of the fermentation process in intact, uncrushed high moisture silage (DLCVGS) is important. The lower costs of the technique merit its study before it is dismissed as having unwarrnated risks.

Animal performance studies are needed to quantify anecdotal reports of excellent palatability, intake and performance.

Better techniques for feeding and methods for reuse of plastic or other pile covers are also needed.

Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the author(s) and do not necessarily reflect the view of the U.S. Department of Agriculture or SARE.