Low Tunnel Materials for Protection and Phenological Acceleration of Figs

Research Plan:

Low tunnels will be setup on rented land at two locations: Lake Drum Orchard, 79 Stark Street, Waterloo, NY, and Anthony Road Wine Company, 1020 Anthony Road, Penn Yan.

The low tunnel setup will consist of 10’ long electrical conduit pipes bent to make hoops that are three feet wide, and about 42” tall. The trees will be spaced eight feet apart, in a 100 foot long row (so 12 trees per row). There will be three tunnel rows at one location (site A), and one tunnel row at a second location (site B). All four tunnels will have interior hoops made from 9 gauge soft vineyard wire that are four feet in circumference (one foot wide). All four tunnels will be covered with 6 millimeter thick greenhouse plastic on the exterior hoops. The interior hoops will be covered with different material combinations that will serve as the treatments for the study. As seen in Table 1, tunnel 3 will serve as the control. In the spring, when the threat of sub 20F temperatures abates, all materials will be removed from tunnel 3. Tunnels 1, 2, and 4 will serve as the separate treatments and will keep their materials in place until growth begins to touch the material. So first the interior material will be removed, then the exterior as growth continues.

Table 1: Tunnel Treatments and Locations

All figs will be trained in a cordon-cane or cordon-spur manner. This requires, and allows, pruning and tying in the fall to fit under the protection scheme of hoops and plastic. Being low and fairly wide creates a large ground footprint under protection. The theory is that ground heat, and not the R-value of the protection materials, is the most important factor in moderating temperature extremes. My familiarity with cordon pruning and growing systems through grapevine production will make phenological tracking easy and reliable.

Three out of four of these tunnels are already planted with fig trees. Six trees were planted in 2018, the other 30 were planted in 2019. This will allow data to be collected regarding fruit production immediately, as 36 out of 48 total trees will be of mature bearing age. Twelve trees will be planted in spring 2021, using the same spacing and will be protected, grown, and trained in the same manner as the older trees. I hope to plant six of the trees to the same variety, and the other six to either another variety or at least the same family. The one downside of the existing plantings is that variety is scattershot. The new planting will help to cut down on some noise. That being said, this study is focused on temperature protection and phenological acceleration regardless of variety.

The temperature probes used for this proposal are made by Ambient Weather/Ecowitt and connect with a sensor hub/monitor via radio wave. The sensor hub/monitor connects to the internet via Wi-Fi and results are available, and recorded, by the minute. There is no local backup of data in the event of Wi-Fi failure. This is potentially problematic but keeps the cost of the project down. Temperature probes will be placed outside of the tunnels, within the tunnel between the interior and exterior hoops, and within the interior hoop if that hoop exists. I will attempt to protect the probes from solar radiation throwing off the temperature during the day by shielding them. The recorded temperature data will be analyzed, comparing lowest, highest, and average temperature experienced. This data can be used to make a growing degree days chart for each treatment, showing the presumed acceleration of ripening through materials used. These probes also measure humidity, so dew point can be calculated. The temperature data from the probe placed outside the tunnels will be corroborated by the closest available weather station.

The material coverings will be applied in the fall when the forecast shows consistent sub-freezing temperatures, or a low under 25F. My guess is early December. The materials will be removed in the spring, from the interior materials to the exterior materials, as the shoots of the tree threaten to touch each layer. If temperatures reach over 95F in the spring, the tunnels will need to be vented by pulling up the plastic on the sides.

Phenology will be recorded first by counting the number of buds left post-pruning and tying to the low cordon system. My goal is 12-15 viable shoots per mature tree. I may leave slightly more, anticipating some damage, then prune those away in the spring with bud emergence. The buds that emerge as viable shoots during bud break will be counted to see how well the protection worked. Any frost damage incurred from advancing bud break through the protection materials will be noted. The date of bud break will be recorded for each tree, in each treatment as well. Next, when figs first form by tree will be recorded. After that, the date on which the first ripe fig is picked will be recorded. Each tree will be tracked by harvest to show total figs per tree over the course of the season. A five to ten fig sample of each variety will be used to find average fig weight so yield weight can be estimated. Pruning weights will be taken in November, before installing the protection materials, to see if the (presumed) phenological boost affected the total growth of each tree. Any unanticipated results that may have a relationship to the protection scheme will be recorded.

Research progress 2021-2022

While my project didn't receive funding until the end of March of 2021 I did have aspects of the project proposal in place prior to funding so I will briefly discuss the results of my protection schemes from winter 2020-2021 as they have bearing on my phenological recording during the growing season of 2021; they also informed my protection scheme for this winter (2021-2022).

Protection 2020-2021

@Site A (Waterloo, NY), 1 tunnel (12 trees)

Clear Poly (6mil) + Clear Poly (4mil) + Row Cover (1.5oz)

@Site B (Dresden, NY), 2 tunnels (24 trees)

North Tunnel- Clear Poly (6 mil) + Clear Poly (4mil)

South Tunnel- Clear Poly (6 mil) + Mylar (2 mil)

The site A tunnel was partially protected on 12/11/20 (outside, clear poly layer only). Site B tunnels were fully covered on 12/23/20 while the site A tunnel was fully covered on 12/24/20. 

Phenology 2021 Growing Season

Budswell was observed towards the end of March and by 4/9/21 a few buds close to the ground had just opened. Despite these early signs, and congruent with the cool April of 2021, the trees really didn’t experience budbreak till close to 5/1/21. And even then they didn’t pick up much steam until about the second week of May. This can be attributed to possible vascular damage from impaired cold hardiness, as well as the cool spring temperatures.

The first fig primordia (bumps before figs emerge from shoot) were observed on 5/31/21 on the early varieties (mostly Mt Etnas like Chicago Hardy). At 6/15/21 the early varieties were at pea size while fig primordial bumps were observed on the later varieties like Galbun, Gros Monstrueuse and Lebanese Red. By 6/29/21 the early varieties were at cherry size, which is very close to lag phase, while the late varieties were at pea size. By 7/9/21 most later varieties had reached cherry size.

On 8/30/21 the first color change was observed on a Chicago Hardy tree at Site B. Color change was then seen on a somewhat later variety (Giant Amber) on 9/5/21 at Site A. Sometime between 8/30/21 and 9/5/21 the first ripe figs were enjoyed by a non-human. As a reference point, the first ripe figs from potted fig trees not associated with this project were enjoyed on 8/7/21. On 9/20/21 the first VdB showed color change. On 10/10/21 the first Galbun was ready for harvest. On 10/15/21 the first AJH ripened (tied with VdB for last to ripen in most years).  

Ideal Tree Case Study 2021

At Site B one tree, a Graye’s Greenhouse (heirloom fig from Mt Etna family), came through the winter with little damage. This tree was observed as an ideal case study on 9/28/21. It had two trunks and one retained sucker to be used as a possible renewal trunk. There were four shoots on one trunk, and seven on the other. From the shoots, including the renewal, 181 figs were counted as being at cherry size or larger. I’m using cherry size as a proxy for the figs being at lag phase. I think 20 figs were already picked by this time, meaning this tree set 201 figs total.

On 11/04/21 this tree was observed as having 83 figs at cherry size left unripened. It stands to reason that between 9/28/21 and 11/04/21 close to 100 figs were picked off this tree.

The full data can be viewed here- https://docs.google.com/spreadsheets/d/1nK3NRFfyOtA3xm4PgP4MBeMI02oc4LrG8mnQXYlbkfY/edit?usp=sharing

Harvest Notes 2021

Around 40-50 figs were left unripened on Gros Monstrueuse. Around 40 ripe figs were picked off four VdB trees. These trees did not benefit from a head start (phenological acceleration) in the spring from the protection materials, in fact they suffered due to the protection materials due to compromised cold hardiness, and struggled to ripen figs in the limited zone 6a season. For anyone planting I’d concentrate on planting early varieties unless the late varieties pull a premium to make up for lack of production. I hope that I’ll see better results (i.e. accelerated growth) this spring with better protection materials.

Pruning Weights 2021

I don’t think too much can be gleaned from this data except that trees that survived the winter better produced more wood in the growing season. The ideal case study tree, Graye’s Greenhouse, grew 5.75 pounds of wood in the 2021 growing season. The average for the site B south tunnel (without Graye’s) was 1.6 pounds, while the north tunnel was 1.48 pounds. The trees at colder Site A, having taken more damage, produced an average of 1.09 pounds per tree.

Full data available here- https://docs.google.com/spreadsheets/d/1nK3NRFfyOtA3xm4PgP4MBeMI02oc4LrG8mnQXYlbkfY/edit?usp=sharing

Tunnel Construction 2021

A new fig tunnel, site B middle, was planted around 6/1/21. It has six trees of one variety, and six that are from the same family (Mt Etna). This uniformity should lend itself to data comparison. All 12 of these trees are early ripening, which should help production. The construction of this tunnel was slightly different than the previous three. Rebar stakes were driven into the ground, and the EMT hoops were placed onto the rebar. The rebar and EMT were joined by EMT couplers. Unlike previous tunnels, the landscape fabric was laid down prior to planting in the spring (much easier than retrofitting). This tunnel, Site B north, and the tunnel at Site A, use a draping wire for the inner protection layer. Site B South uses small hoops made from nine gauge vineyard wire. All tunnels now have a “purlin” wire that runs through the EMT hoop at its apex and is secured to the end posts. This prevents the hoops from wiggling free from where they’re attached to their stakes.

Unfortunately, lack luster planning and a late grape harvest season pushed back the completion of this tunnel until 12/27/21. Protection was put into place at all tunnels by 12/27/21, but was not fully buttoned up until 1/7/22. Supply chain issues delayed the placement and activation of temperature sensors until 1/8/22 at Site A and 1/15/22 at Site B.

Protection 2021-2022

Site A- Clear Poly (6 mil) + Row Cover (1.5oz)

Site B, South- White Poly (4 mil) + Mylar (7 mil)

Site B, Middle- Clear Poly (6 mil) + Mylar (7 mil)

Site B, North- White Poly (4 mil) + Clear Poly (4 mil)

This setup will allow us to decouple the influence of the white poly, outer, and mylar, inner, layers. Unlike the mylar from the 2020-2021 season, this mylar is white on one side. The reflective side was faced inward, to retain the ground heat. We’ll see how the white outside layer does in reducing heat spikes from the winter sun.

2022-2023 Disasters and Pivoting

During the winter of 2021-2022 voles damaged or destroyed 90% of the fig trees overall, and 100% at the Dresden location. Around 1/3 of the trees in Dresden were not salvageable and had to be replaced in the spring of 2022. The remaining 2/3rds were set back to the soil line due to girdling. Some trunks may have suffered winter damage in addition but it was difficult to decouple the two effects. My best guess is that the cozy conditions of the low tunnels caused first a slow build up (the first few winters) then an explosion of the vole population (2021-2022 winter). The sheer numbers of the voles prevented their control by poison alone, and encouraged feeding on the fig trunks due to competition and food scarcity.

Some relevant data was collected by the temperature sensors which will be discussed in the results section, but from a farming perspective the voles effectively killed any hopes for growing season research--recording pruning weights, phenology dates, or crop weights and counts. With this in mind I pivoted this project to paying more attention to vole protection, and a discussion of what went wrong. 

Landscape Fabric

After the full scope of the destruction caused by the voles became apparent in the spring my first move was to remove the landscape fabric. My original idea forthe fabric was to warm up the soil, accelerating phenology, and keep the weeds at bay since as a side project I'm not always able to devote as much time to the figs as I'd like. Unfortunately, it proved an ideal habitat for the voles. Upon removal I found extensive tunneling, often leading up to, under, and through the roots of the fig trees. I left landscape fabric on one fig tunnel row (out of the four total) as a "high pressure control" of sorts, to see if my other vole damage prevention measures are enough on their own. In the 2022 growing season weeds were controlled with a combination of weed whacking and contact (non systemic) herbicides. The landscape fabric did seem to encourage a strong native snake population, but unfortunately they aren't active during the winter when we need their help.

Vole Protection 2022 Edition

Previously my vole protection consisted of PVC bait stations filled with zinc phosphide bait, generally two per 100 foot tunnel, and hardware cloth around at least 50% of the trunks for each tree. After watching the voles dig under the hardware cloth, first eating through the roots and then emerging to girdle the trunk, I decided to ditch this method entirely. I do think it could be effective if you are able to place the hardware cloth deep enough at the same time as planting the tree--but it would need to be wide enough to accommodate future growth. In the late summer of 2022 I started overhauling the vole protection program by first digging around the root ball of each tree. I removed as much dirt as I could then back filled with no.2 stone. The idea being that the voles would be discouraged by trying to dig through the stones and find easier meals. That left above ground protection, which I attempted to solve by painting the trunks with a paint and castor oil mix. I used interior latex as it's supposed to be both persistent and permeable. Castor oil is supposed to be a strong rodent repellent. I did not dilute the paint and the ratio was something like 5:1 paint to oil (this is significantly more oil than recommended but their use cases weren't with paint). A few extra trunks were left unpainted as a poor man's control for this experiment. And again one tunnel at the Dresden location still has landscape fabric so it's the "high pressure control" of sorts. The bait stations will be refilled more aggressively over the winter of 2022-2023 than they've been in the past. 

2022 Changes to Low Tunnels

A few changes were made to the low tunnels at both locations in preparation for the 2022-2023 winter. The clear poly used for the 2021-2022 winter at Dresden was 2 mil overwintering clear plastic instead of the intended 6 mil clear plastic. This was a bone headed mistake and will be discussed further in the discussion section. So this winter the correct 6 mil clear poly is in place. No row cover will be used this year at the Waterloo tunnel location. Instead, a double outer layer (white 2 mil overwintering plastic, then 6 mil clear poly inside) is in place, with an inner layer of mylar.

At the Dresden location the north most tunnel suffered extensive wind damage in the fall when covering materials were left attached to the ends but not secured to the individual hoops. In response, I took down all the hoops and am instead calling the north tunnel the low-low tunnel as it's a 20 inch tall triangle shape (the regular hoop tunnels are 42" tall). This tunnel has two layers: white 2 mil overwintering plastic, then 6 mil clear poly. Both layers have been doubled up by folding them over top of themselves, and secured using tarp clamps and vineyard wire. If temperature data shows this method to be beneficial it's an attractive and simple alternative to the EMT hoops which require several onerous steps before and during setup. 

Misc Notes: Irrigation stakes (1/2") are a cheap and useful option for anchoring the plastic to the ground, especially next to where the rope tie downs are secured as that creates a natural gap. The tunnel ends have always been tricky to secure and I've found spring clamps to be simple and effective. In the summer of 2022 I left some of the covering materials in the field by securing them to the purlin wire that connects the EMT hoops. They didn't suffer any significant damage though they would sometimes fill with rain water (I found tying them tight together to the wire at least once between the hoops to help cut down on this). It did make the fall covering easier, though it likely shortens the lifespan of the covering materials.