Research Article Review: Early almond harvest as a strategy for sustainable irrigation, pest and disease management

Citation: Evie E. Smith, Patrick H. Brown, Ellie M. Andrews, Kenneth A. Shackel, Brent A. Holtz, Daniel J. Rivers, David R. Haviland, Sat Darshan S. Khalsa, Early almond harvest as a strategy for sustainable irrigation, pest and disease management. Scientia Horticulturae, Volume 293. 2022. https://doi.org/10.1016/j.scienta.2021.110651Early harvest has been a strategy to mitigate navel orangeworm (NOW) infestation while assisting with harvest operations. Most almond farmers define early harvest as when the hulls are split, with some yellow hulls still drying within the tree. This is often around stage e in the UC IPM hull-split guide (figure 1). (Please see here for all of the tables and figures)A recent paper by Smith, et al, 2022 reviewed the effects of starting almond harvest earlier. Harvesting earlier has the potential to reduce  (NOW) infestation, hull rot canopy losses, and potentially aid in operations while reducing water demand of the tree. This experiment was established within two farms: one near Woodland, California which was a 2015 planted organic block, irrigated with double line drip, and being mechanically harvested for the first time in 2020. The second orchard was a mature orchard near Denair, California which was 11 years old, irrigated on double line drip, and had minor problems with NOW, and some issues with hull rot. This experiment duration was one year.The experimental design consisted of a randomized complete block design with four blocks of each treatment. Individual tree samples consisted of three trees within each block, or about 24 trees total across both treatments for each location. Early harvest (EH) was defined as removing fruits 3-4 weeks earlier than standard practice.  Within the Denair site, EH and standard harvest (SH) occurred 22 days apart, on July 24th and August 14th, respectively. At the Woodland location, the harvest difference was 27 days, with EH and SH occurring on July 30th and August 26th, respectively.To assist with harvesting, EH treatments were deficit irrigated between late May and July. Trees received water based on the use of the pressure chamber, in which mid-day stem water potential was maintained between -15 and -18 bars. The SH received the grower standard. Additionally, several other variables were measured, including shake timing, kernel harvest characteristics, yield, the amount of NOW infestation and hull rot, and total water applied. It should be noted that kernel moisture was dried to 3% across treatments prior to calculating yield and kernel characteristics.The researchers found that the EH and the associated RDI scheduling of irrigation led to a reduction of water applied to the orchards. Within Denair, EH and SH had water applications of 21.2” (536 mm) and 24.4” (620mm) per acre, respectively. At Woodland, a similar trend was observed, in which 19.2” (488 mm) and 20.8” (528 mm) per acre were applied to the EH and SH treatments, respectively. As a percentage, EH received 86% and 92.4% of the SH irrigation at Denair and Woodland sites, respectively.There were some differences in the phenology of the crop at the time of shaking (table 1). These differences make sense as EH started earlier, leading to less time for the fruit´s natural drying progression. Yield between the two treatments was not influenced (table 2), even though differences were observed numerically. EH yielded 83% and 92.5% of the SH treatment at the Denair and Woodland, site, respectively. Interestingly, this is close to the percentage of water reduction that occurred when RDI was integrated into the EH treatments. Pest and disease occurrence differed amongst the locations (table 3). There were no observed differences in hull rot. NOW infestation was lower at the Denair site, but not the Woodland location. Return bloom wasn’t counted consistently across the trial locations. The authors concluded that the potential benefits of EH were evident within the study, and advocated for greater experimentation and possible utilization of the practice.Although I found this paper interesting, I felt that the major concerns about harvesting almond early weren´t discussed. First, it is important to note that the yield differences were most likely NOT due to the shake timing, but rather the water deficit that was required in order to shake earlier. Previous research has shown that in-season water reductions reduce yield on a 1:1 basis, meaning that a 1% reduction in water will lead to a ~1% reduction in crop. This is similar to the yield reductions observed in the experiment. If RDI is required to harvest almonds earlier, than yield reductions should be expected.Shaker damage was not measured or observed between treatments. Shaker damage is often higher in earlier harvested almonds due to the bark tightening as the season progresses. This variable wasn’t measured as trees were individually harvested. Considering that fungal infections of wounds caused by shaking are one of the primary causes of early orchard decline, this was a bit disappointing. Additionally, what is concerning, is the increase in shaker timing. EH trees required a 5-6 second shake to achieve good removal, while SH required a 3 second shake. This would have a marginal increase on shaking expenses, as well as lead to more trunk damage.One thing that wasn’t clear was the difference between kernel yield and marketable yield. The researchers indicated higher marketable yield within the SH treatments, but didn’t identify what this meant. It does make sense, however, that starting harvest too early may impact marketability of almond kernels. Nuts harvested too early may not fully develop, and have increasing frequency of mild kernel shrivel, or “texturing.” Personal experience has found that this tends to occur more frequently as nuts are harvested in stage B3 or earlier.It is unsure how easily it will be to shift to a greener harvest. Shaking nuts earlier would require more drying time, slowing harvest, and creating risk exposure to early rains. Additionally, it may lead to greater amounts of ant damage. Removing the crop too early from the field will require the added expense of dryers, especially for later varieties which will dry slower on the ground. This may increase harvesting expenses as dryers are expensive to operate due to the use of natural gas.My experience in the initiation of shake timing is that most California operations could start earlier. I have observed that a good shake when starting once the tree is 100% at stage b3 or c. This should be based on the wettest part of the field, and on the nuts within the internal part of those trees. This will often accelerate shaking by 1-2 weeks in comparison to normal practice, and not require so much drying time. Shaking at this timing may also reduce the need to re-shake as internal nuts will be mature, but still heavy enough to shake free.I believe that the greatest benefits of starting earlier are tied to operations and nut removal. Starting earlier would allow more days for harvest operations, reducing equipment needs on the farm. This will be more important as farms integrate solid plantings of self-pollinating blocks, or within operations that have a large area of a single variety (e.g. Nonpareil). Nut removal for some varieties also tends to be improved with earlier harvest. For example, shaking varieties that tend to fruit on the primary and secondary scaffolds (e.g. ´Independence´, `Wood Colony´) can lead to greater shaker efficiency. These practices, if properly integrated can improve harvest efficiency.Early Harvest Tables and Figures