Final report for FNE24-095
Project Information
The purpose of this project was to determine whether changing garlic planting methods from 4-rows per bed (current standard) to 3-rows per bed would result in improved yields or larger bulbs, thereby optimizing production and profitability for small farms.
To test this, six study beds were planted in October 2023, with three beds containing 3-rows and the other three containing 4-rows. Row lengths were standardized for analysis, and all rows were treated equally regarding fertilization, mulching, insect netting, and weeding. The garlic was harvested, segregated, and analyzed by weight, bulb size, and the number of bulbs >2 inches in diameter. Statistical analysis was performed to compare the results using a paired-comparison method.
The study found that the 4-row configuration resulted in significantly higher gross yields by weight (primary outcome measure) compared to the 3-row configuration. However, there was no statistically significant difference in the number of bulbs per row or the number of bulbs greater than 2 inches in diameter. The findings indicate that switching to a 3-row method is not economically or agronomically advantageous, as it does not lead to a meaningful increase in high-value seed-grade garlic and requires a comparable amount of planting effort.
This research was particularly informative for our farm, confirming that the 4-row method is optimal under our specific conditions, which include nutrient-rich organic no-till beds with heavy composting. While the results may have limited applicability to other farms with different conditions, they provide a valuable baseline for further exploration, such as testing 5-row configurations.
This project seeks to compare the agronomic and economic effects of 2 row spacing configurations for garlic production. Specifically, our objectives are to determine if the following response variables differ between row-spacing configurations:
- Yield
- Bulb size
- Percentage of high-value bulbs (> 2” diameter)
Additionally, we will perform an analysis of net returns as a function of garlic row-spacing to determine the most economically viable system.
THE PROBLEM
Our farm is small, every square foot matters. Maintaining an operation that is financially sustainable means that we must continually optimize and improve. We research existing approaches, experiment with new methods and tools, seek the counsel of experts, and document all of our activities. One of the ways we improve our production and profitability is by increasing the average size of our garlic bulbs. This is important for us for two main reasons: 1) we sell our garlic by weight; 2) the largest bulbs (>2” diameter can be sold for 20% higher price as seed garlic, and; 3) like other small farms, the area we have under production is limited and the potential for expansion is constrained by cost and availability.
While we have been selecting the largest bulbs over the past 4 growing seasons, this has been the only method of potentially improving our net returns given our farm’s size. Given this, we are interested in the effects of row spacing on garlic bulb size. Specifically, we would like to answer the following question: Which spacing configuration is economically superior, 3-rows/bed (2.2 bulbs/sqf) or 4-rows/bed (3 bulbs/sqf) considering bulb weight (since larger bulbs yield a higher price) and garlic bulb-weight per linear foot of bed space?
THE SOLUTION
Determine whether another spacing configuration, which is feasible to implement, will yield economically-superior results for our farm - and potentially other farmers in our region, growing similar varieties. In 2021 and 2022 we dibbled our beds by hand, with a dibbler that made six holes at a time, each spaced 6” apart (in-row). This method was prone to inconsistency and time consuming. In 2023 we purchased and began dibbling our beds with a tow-behind implement consisting of a set of wheels (each weighing 35 lbs) which spaces each dibble approximately 6.5” apart (in-row) with adjustable (the wheels can be moved along the axel) between-row spacing. Thus, we have the ability to test 4-row/bed vs. 3-row/bed spacing with efficacy. This is the experiment that we have initiated (while we have started the project prior to March 1, 2024 - a necessity due to constraints of our growing season for garlic in the Northeast - we have not requested funds to pay for any expenses that will have taken place prior to March 1, 2024).
SIGNIFICANCE
The potential of this project to inform growing practices of garlic growers in the Northeast is significant. This research is particularly relevant for 1) growers of seed garlic, 2) growers with limited land, and 3) garlic growers using no-till methods and/or permanent beds. If, for example, the study demonstrates the potential that 3-row spacing (in a standard 30” permanent bed) results in significantly larger (based on weight) bulb size, growers would have more bulbs to sell as seed garlic at a price of $25-$28/pound vs. $18-$20/pound for culinary garlic. This pricing variation-benefit could offset a loss of production due to wider spacing. Conversely, many growers currently plant 3-rows of garlic per bed. If the data indicates the net benefit of planting 4-rows/bed (compared to 3-rows) is significant, growers would have the evidence they need to convert their spacing strategy in their permanent beds with confidence.
Saunderstown Garlic Farm has been operating, commercially, for 4 seasons. We are a certified-organic farm and utilize no-til practices. We grow only garlic and (we are fairly certain) are the largest garlic farm in Rhode Island. In 2024 our gross sales were about $18,000. We are a part-time operation.
Cooperators
- - Technical Advisor
Research
The project/research began in October 2023 when we planted six study-rows (three groups, each consisting of one bed with 3-rows of garlic seed, and one bed with 4-rows of garlic seed). The length of each row was measured (since there was variation in row-length), and data was normalized to a standard (1,000 bed-feet) during analysis. The study-rows were planted after our non-study rows were planted to reduce the possibility of error and/or contamination (our planting day involves 20-30 volunteers and can be hectic). Post-planting, the study-rows were covered with straw mulch (the same as our other non-study rows), and labeled. No beds (neither study or non-study beds) were irrigated until early May.
In early March, at which time we covered all of the study rows with insect netting, to safeguard the crop against Alium Leaf Miner (ALM). While ALM has had limited impact in fall-planted garlic crops in Rhode Island, to date, there is not yet consensus in the growing/academic community (per Dr. David Weisberger, URI Extension) about how to best mitigate (or even if mitigation efforts are necessary) against ALM. We decided to cover the study-crop (as described in our application) out of an abundance of caution, to ensure that we did not have one more variable (ALM) to account for.
In March, and again in early April, all study-rows were amended with nitrogen (sodium nitrate) at a rate of 40 lbs per acre.
In June, scapes were removed from all study-rows. Scape removal was performed approximately twice per week for the month. Scapes are removed when they have fully curved. At each scape removal-event, scapes (that were ‘ready’) were removed. Throughout May and June, it was also necessary to weed the study rows, which took more time than normal because, even though we use mulch to reduce weed pressure, some of our vetch cover crop (planted September 2023) was terminated too late, and seeded in the study-rows. The weed pressure in each of these study-rows was materially similar, and when weeding occurred, all rows were weeded in the same day.
In July, approximately 50 individuals, over two days, harvested our garlic crop. A select group of these folks were briefed on the research project, focusing on the importance of segregating and properly labeling the garlic bulbs harvested from each study row. The crop from each study-row was segregated for cleaning (removal of stems and roots), placed in individual bins (by row #), and placed on tables in the high tunnel for curing. Garlic resulting from each study-row were also segregated, and appropriately labeled, and placed on perforated metal benches.
After approximately 10 days of curing, Dr. Weisberger joined farm representatives to count, weigh, and grade (less than and greater than 2” in diameter). All data was recorded.
In late October, in partnership with the URI Extension office and the Young Farmers Network, we held an event at the farm during which we discussed the results of the study. During this event we also shared, more generally, information about our approach to organic farming/growing garlic. Refreshments were served and the event was attended by about 20 people.
SARE-SGF Garlic Study Analysis - _2_ bulbs
SARE-SGF Garlic Study Analysis - Number of Bulbs
SARE-SGF Garlic Study Analysis - Weight
As articulated in our application, our primary outcome measure was total bulb weight per row, with secondary measures: number of bulbs per row, and number of bulbs great than two inches in diameter. Our aim was to determine whether the differences between study/control rows would be significant enough to warrant changing our standard planting-method from 4-rows of garlic per bed (standard), to 3-rows of garlic per bed. Based on these results, we would (if appropriate) conduct further sensitivity analysis (economic) to determine if changing our standard approach to planting (from 4-rows to 3-rows/bed) would be economically-prudent (based on both potential revenue change, and change in labor). The reason we included the bulb size as one of our measures is that, typically, garlic that is greater than 2 inches in diameter can be sold as seed garlic, which commands a price that is about 20-25% higher than culinary (less than 2” in diameter) garlic.
With the help of Dr. Weisberger, we conducted a simple statistical analysis, based on the “paired comparison” method, as described by S. Gailans (2014) (see full citation below). The raw data, and our computations, are attached as a separate PDF. As these results indicate, analysis related to the primary outcome measure yielded a statistically significant result (95% CI). Thus, we have concluded that the 4-row method (planting 4 rows of garlic seed per bed), which is our standard method, is better than the 3-row method when considering gross yield by weight. The analysis associated with the other two measures (number of total bulbs per row, and number of bulbs > 2” per row) did not yield statistically significant results.
Our takeaway from this research is that changing our planting methods is not warranted from either an economic or a time perspective (given that the amount of time required to plant a bed with 3-rows of garlic vs. 4-rows is not materially different). If our results had indicated that 3-row beds yielded a statistically significant increase in 2” inch (seed grade) bulbs, we would have performed additional analysis to consider whether that increase was enough (given the higher price that can be demanded for seed garlic) to justify switching to 3-rows EVEN THOUGH the 4-row treatments yielded higher gross weights.
This study thus was informative and helpful to us, reassuring our team that reducing the number of garlic that we plant per bed will not lead to a better result.
OBSERVATIONS/DISCUSSION
This study was informative for us, but it’s utility to other garlic growers is likely limited to those growing in similar conditions. The primary condition that we feel may have influenced our particular results is that we plant in a nutrient rich environment. Our farm follows organic (and is certified-organic), no-till methods, using cover cropping and rotating crop-fields. Most notably, our study beds consisted of 4 inches of compost, which was spread only two months prior to planting. We theorize that a primary reason that we did not observe a significant difference between the 3- and 4-row/bed treatments is because the seed/bulbs have sufficient nutrient resources and did not have to compete with each other. This begs the question: What is the optimal number of rows (based on the square-footage of bed space)? Could, for example, 5 rows (which would require spacing the seed at significantly less than the recommended # of seeds/square-foot) yield a gross yield (weight) that was significantly indifferent than 4 rows? This is something we may explore in the future.
- Gailans, “Two-Treatment Trial Designs for Conducting on-Farm Research,” 2014. updated from “The Paired-Comparison: A Good Design For Farmer-Managed Trials” by R. Exner and D. Thompson, 1998.
As described in the previous section, based on the results of this study, we will not be making changes to our in-row spacing. This study was very useful and we are appreciative to SARE for the opportunity to dedicated the time and resources to answer this key question - which has important implications for our small farm, and we hope other garlic growers will also find it informative. We were able to answer our research questions, and as indicated in our analysis, successfully assessed the outcome measures that we described in our application.
Education & Outreach Activities and Participation Summary
Participation Summary:
As described in our main report, we hosted a field day on October 30, 2024, at which we presented results of the study and discussed more generally the farm's operations.
Learning Outcomes
Optimal spacing for garlic seed.
Project Outcomes
This study confirmed that our current approach of planting (at least) 4-rows of garlic per 30" bed is optimal, compared to 3-rows (or less).
See main report for additional details.
Our research methods were simple, and worked well. We are curious to know what the threshold is for the number of rows of garlic seed can be planted per 30" of heavily-composted bed, without sacrificing yield.
I think this study would benefit other garlic growers who use organic, no-till methods, who also follow similar methods to our farm - namely no-till/cover cropping and planting in at least 3" of compost.