In the last issue of this UA Vegetable IPM Newsletter, I presented a melon (Cucumis melo ‘reticulatus’ L.) crop phenology model (Figure1; Silvertooth, 2025). This model can be useful in predicting and tracking crop development and identifying important stages of crop growth and development (crop phenology).
Use of a crop phenology model can be applied to basic crop management (e.g. fertilization, irrigation, harvest scheduling, pest management activities, labor, and machinery management, etc.). For use of the crop phenology model, good local weather data with heat unit information is needed. In Arizona we have an excellent weather system with the Arizona Meteorological Network, AZMET.
Since cantaloupes are a warm season crop, we use the86/55 ºF heat unit (HU) thresholds for phenological tracking. Key stages of growth or “guideposts” indicated in Figure 1represent the average or “target” values that are subject to a slight degree of natural variation, which is normal.
This melon crop phenology model was developed under fully irrigated and well-managed conditions. That is important since non-irrigated fields are more likely to experience water stress, which significantly disrupts crop development patterns.
Referring to the data from AZMET for several locations in the Yuma area, the HU accumulations (86/55 ºF thresholds) from 1 January 2025 to a set of four possible 2025 planting dates are listed in Table 1. The HU accumulations from 1 January2025 to 15 April 2025 for these sites are listed in Table 2.
The HU accumulations after planting (HUAP) for these four possible planting dates for three Yuma area locations to 15 April 2025 are shown in Table 3. The HUAP values in Table 3 are simply the difference between the values in Tables 1 and 2. An example for the Yuma Valley, 15 January 2025planting date is: HU - 73.1 HU = 645.6 ~ 646 HUAP.
The information in Table 3 can help serve as a reference to check for melon crop development in the field against this phenological model in Figure 2. Several cantaloupe/melon types are being in this region including western shipper type melons, Tuscan melons, and Hami melons. In the past, each of these melon types have tracked closely with this phenological model.
For melon crops in the lower Colorado River Valley at this time, we can expect to find fields planted and watered up in mid-January to have crown set melons beginning to develop netting, which may not be as apparent on the Hami melons. These fields could have crown fruit ready for harvesting in about three weeks, based on normal HU accumulation patterns for this time of year. For fields planted and wet dates near the first of March, these fields should be vigorously flowering.
Reference:
Silvertooth, J.C. 2025. Tracking Cantaloupe (Melon) Crop Growth and Development. University of
Arizona Vegetable IPM Newsletter, Volume 16, No.7, 2 April 2025.
Table 1. Heat unit accumulations (86/55 ºF thresholds) after 1 January 2025 on four possible 2025 planting dates utilizing Arizona Meteorological Network (AZMET) data for each representative site.
Yuma Valley: https://azmet.arizona.edu/application-areas/heat-units/station-level-summaries/az02
Yuma North Gila: https://azmet.arizona.edu/application-areas/heat-units/station-level-summaries/az14
Roll: https://azmet.arizona.edu/application-areas/heat-units/station-level-summaries/az24
Table 2. Heat unit accumulations (86/55 ºF thresholds) after 1 January 2025 to 14 April
2025 utilizing Arizona Meteorological Network (AZMET) data for each representative site.
Table 3. Heat unit accumulations (86/55 ºF thresholds) after planting (HUAP) from four
possible 2025 planting dates and three sites in the Yuma area on 15 April 2025 utilizing
Arizona Meteorological Network (AZMET) data for each representative site. Each value
is rounded to the next whole number. Note: the values in Table 3 are determined by
taking the difference between the HUs foreach representative site and four planting
dates in Tables 1 and 2.
Figure 1. Melon (cantaloupe) phenological development model expressed in Heat Units
Accumulated After Planting (HUAP, 86/55°F).
First, I want to thank everyone who participated in last week's Vegetable Pest Losses Survey.
This year's survey included the return of the lettuce disease losses section. While several diseases were present and managed last season, downy mildew accounted for the majority of disease management costs by a wide margin. This really underscores the
impact that last spring's unusually rainy weather had on disease development across the Yuma lettuce production region.
No one can predict exactly what this upcoming spring will bring, but there has been discussion about the possibility of a strong El Niño leading to an extended monsoon season. If that proves true, the conditions would once again support spring
downy mildew development. The pathogen only needs about 3 to 4 hours of continuous leaf wetness to infect lettuce, so periods of overnight moisture, prolonged morning dew, or frequent rainfall when inoculum (spores) are present increase disease risk.
With that in mind, this seems like a good opportunity to review what is known about downy mildew and discuss strategies for its management.
Resistance in lettuce to Bremia lactucae, the causal oomycete pathogen behind downy mildew, is inherited in a gene-for-gene fashion, meaning one major gene product in the plant host interacts with one major gene product in the pathogen. When
resistance is present, this leads to an incompatible interaction between plant and pathogen and results in complete immunity to infection. Resistance genes in these types of interactions most often encode a protein molecule that acts like a burglar
alarm. These molecular sensors in the host bind to proteins secreted specifically by the pathogen, and when that happens a storm of defense responses is activated in the plant that excludes further infection. This is not the only mode of genetic resistance
found in plants, but it is often the most drastic and effective against obligate parasites like downy mildew.
But this simple gene-for-gene interaction often puts incredible selection pressure on the pathogen populations to change such that they can get around the resistance. By losing the molecule that the plant detects in order to initiate a defense response,
the pathogen becomes unrecognizable to the resistance genes a plant variety may have. Just like spraying the same mode of action over and over again leads to a pest population developing resistance to a pesticide, the same selection applies to genetic
resistance. The longer a resistance gene is deployed in a region, the more likely the pathogen population is to change in response until that resistance gene is no longer effective at managing the disease.
One of the biggest challenges with lettuce downy mildew is that B. lactucae is constantly changing over time. It exists as many different races, where each race has a different reaction to the resistance genes bred into lettuce varieties. You
can think of these races as different versions of the same pathogen. A lettuce variety that resists one race may still be susceptible to another.
These races are identified by testing them against a panel of lettuce varieties with known resistance genes. In the western United States, races are named by the International Bremia Evaluation Board-U.S. (IBEB-US) and are given names with a number followed by the country’s abbreviation, such as 8US, 9US, or 10US. The populations found in the western U.S. are different from those found in Europe, so each region uses its own independent naming system.
The downy mildew population has changed considerably over time. Earlier races (1US through 4US) are now rarely found in commercial lettuce production. During the 2000s and 2010s, races 5US through 8US became the most common. Race 9US was recognized after being detected repeatedly between 2015 and 2017, and the newest officially recognized race, 10US, was designated in 2025. Below is a pie chart showing the relative frequency of the races identified from 114 Yuma County downy mildew samples between 2023-2024:
Figure 1: Pathotyping results of 114 lettuce samples from Yuma County collected between 2023 and 2024. Data source: https://bremia.ucdavis.edu/bremia_database_main.php
The results show that much of the downy mildew population found in Yuma County is made up of novel strains of Bremia lactucae that have not yet been officially classified as a race. An official race is only recognized after it has been shown to be stable and widespread over multiple locations and growing seasons. These newer strains may disappear over time, or they may eventually become established and earn an official race designation. In the meantime, this means growers and lettuce breeders in Yuma County are often dealing with strains that can dodge the resistance in some lettuce varieties before those strains are common enough to be officially recognized. It also highlights why relying on resistance alone is not enough to manage the disease.
Table 1: Pathotyping and fungicide sensitivity results of samples from Yuma County collected in 2025.
This trend appears to be continuing. All of the downy mildew samples sent for race testing last season were identified as novel strains rather than known, officially designated races.
It's impossible to predict exactly how these new strains will respond to the resistance genes found in today's commercial lettuce varieties. However, because they have not been previously characterized, they are more likely to overcome existing genetic
resistance than the races we already know about.
New strains develop naturally over time. They can arise when different strains exchange genetics (i.e. intermate) or through random mutations. When growers plant varieties with similar resistance packages over large areas, the pathogen population
is placed under strong selection pressure. Any strain that happens to acquire the ability to infect those resistant varieties gains a major advantage and gets to reproduce without competition where other strains cannot. Over just a few disease
cycles, those successful strains can become much more common in the population until they are the predominant strain overall.
An important point to remember is that the resistance bred into commercial lettuce varieties is not wearing out or becoming weaker over time. The genetics in the lettuce remain just as effective as when the variety was released. What changes is the
pathogen. As the downy mildew population evolves new strains emerge that can bypass resistance genes that previously worked very well.
That means that varieties carrying resistance to races 5US through 10US are still doing exactly what they were designed to do. They continue to suppress those known races and help prevent them from becoming widespread in commercial fields. So, if
you experience significant downy mildew in a field planted with a variety that has a strong resistance package, the culprit is most likely one of these newer, uncharacterized strains rather than a failure of the variety itself.
Unfortunately, Bremia lactucae can evolve much faster than scientists can identify new races and breeders can develop and release resistant varieties. That's why no resistance package should be viewed as a stand-alone solution.
This is also why extension, researchers, and the seed and crop protection industries place so much emphasis on the integrated pest management (IPM) concept. Genetic resistance is an essential tool, but it works best and remains the most sustainable when combined with other management practices. For novel strains that can slip past host resistance, timely fungicide applications and other disease management strategies become especially important for maintaining control.
Figure 2: Mean disease severity by treatment. Disease severity was determined by rating 10 plants within each of the five replicate plots per treatment using the following rating system: 0 = no downy mildew present; 1 = downy mildew present on bottom leaves of plant; 2 = downy mildew present on bottom leaves and lower wrapper leaves; 3 = downy mildew present on bottom leaves and all wrapper leaves; 4 = downy mildew present on bottom leaves, wrapper leaves, and cap leaf; 5 = downy mildew present on entire plant. Disease severity is displayed as the mean of five replicates across all three lettuce varieties and bars show a 95% confidence interval around the mean calculated from individual treatment data. Compact letter display (CLD) above boxes show significantly different treatments (Kruskal-Wallis ANOVA and Dunn’s test). Boxes sharing the same letter(s) are not significantly different from one another. Fb = “followed by” in the rotation. Not all products are registered yet for use in lettuce. The inclusion of specific fungicide products or formulations in these trials does not constitute an endorsement or recommendation over other labeled products.
The most effective way to manage lettuce downy mildew is to use an integrated approach. Plant varieties with a strong resistance package against races 5US through 10US, and pair that resistance with timely, full-label-rate fungicide applications when environmental conditions favor disease. This combination provides the broadest and most reliable protection against both known races and the novel strains that continue to emerge in Yuma County.
If you have any concerns regarding the health of your plants/crops please consider submitting samples to the Yuma Plant Health Clinic for diagnostic service or booking a field visit with me:
Christopher Detranaltes, Ph.D.
Cooperative Extension – Yuma County
Email: cdetranaltes@arizona.edu
Cell: 602-689-7328
6425 W 8th St Yuma, Arizona 85364 – Room 109AI is in the news all the time being touted as the most influential human innovation in history. Why? I’m not sure exactly as generally few specifics are given, but recently came across an article that gave me some insight as to what the experts are talking about. The piece focuses on a project by Google DeepMind1: two AI robots playing soccer (Paul, A., 2024). There’s a couple of things that struck me most about the robots. The first was how human-like their behavior and actions were. They reminded me of a couple of 5-year-olds playing soccer. The second is the manner in which they learned. Unlike traditional programming methods where every action is meticulously coded, these robots were given the objective of scoring a goal and only provided instructions on how to stand up and kick a ball. The rest they taught themselves using deep reinforced learning (AI) methods. The news program 60 Minutes also reported on the AI soccer playing robots (Fig. 1). Check out the article here and the 60 Minutes segment in the video below. I think you’ll be impressed.
Fig. 1. 60 Minutes Segment: Google DeepMind demos AI training robots to play soccer / football. (Credit: 60 Minutes).
References
References
Paul, A. (2024). Watch two tiny, AI-powered robots play soccer. Miami, Fla.: Popular Science. Available athttps://www.popsci.com/technology/deepmind-robot-soccer/.
____________________
[1] Reference to a product or company is for specific information only and does not endorse or recommend that product or company to the exclusion of others that may be suitable.
Today, the EPA posted in the Federal Register an Emergency Order suspending the Registrations of All Pesticide Products Containing Dimethyl Tetrachloroterephthalate (DCPA). We will include the link to the official document at the end of the article.
The notice says in the II. Emergency Order paragraph the following:
“Effective immediately, no person in any state may distribute, sell, offer for sale, hold for sale, ship, deliver for shipment, or receive and (having so received) deliver or offer to deliver to any person any pesticide product containing DCPA. Additionally, in accordance with FIFRA section 6(a)(1), EPA has elected not to permit the continued use of existing stocks, consistent with its policies applicable to cancellations where the Agency has identified significant risk concerns. See 56 FR 29362, 29367, June 26, 1991 (FRL-3845-4)”.
Also, the same paragraph in the document states clearly: “Accordingly, this Emergency Order expressly prohibits any person from using any pesticide product containing DCPA for any purpose. However, EPA will allow continued distribution of existing stocks of DCPA for the express purpose of returning any DCPA product to the registrant of such products”.
You can find and download the document posted in the journal today following this link:
https://live-azs-vegetableipmupdates.pantheonsite.io/sites/default/files/2024-08/240807_EPA_DCPA_ORDER_2024-17431.pdf
References:
We have conducted a field efficacy trial evaluating the efficacy of 14 biological insecticides alone or as a tank mix against lepidopteran pests, including diamondback moth (DBM), beet armyworm(BAW), and cabbage looper (CL). The insect pressures were relatively low when we initiated the insecticide applications; the CL number was never high enough to be considered for statistical analysis and treatment comparisons.
We applied all insecticides at the highest label rate when sprayed alone or at mid-rate when sprayed as a mixture of two insecticides using an application volume of 40 gal/ac. The adjuvant, Oroboost, was added to each of the insecticide treatments at a rate of 0.4% v/v. Most of the insecticides evaluated in our trial are registered for lepidopteran control except for M-Pede, BotaniGard, and PFR-97.
The results of our trial showed that Xentari, Xentari + Pyganic, and Entrust provided the highest level of BAW suppression. We also found that other insecticides/mixes, including Aza-Direct, Dipel, Dipel +Pyganic, Gargoil, Grandevo, Venerate, M-Pede, and PFR-97, may also cause some levels of BAW suppression (Figure 1A). Xentari and Dipel + Pyganic provided the best DBM suppression, followed by Xentari + Pyganic, Dipel, and Entrust, which provided 50-60% of DBM suppression (Figure 1B). Pyganic alone did not control either BAW or DBM (Figure 1A&B).
Table 1. List of bioinsecticides evaluated
Figure 1. Means Beet armyworm larvae (A) and Diamondback moth (B) per cabbage plant as affected by bioinsecticide sprays.
Organic farming faces two major challenges: weeds, which remain the number one concern, and insect pressures that can severely affect crop quality and yields. Effective strategies for managing weeds and insects are critical, especially as organic production expands. Traditionally, these tasks have been labor-intensive and time-consuming, creating a strong need for innovative solutions that can improve efficiency while maintaining crop standards. With the current momentum for increased AI integration into agriculture supported by both industry and Washington D.C., novel technologies AICropCAM present exciting opportunities for leafy green growers. These high-tech platforms could offer early-stage detection of weeds and insects, helping growers respond more quickly and precisely. Such innovations have the potential to save significant time and labor, particularly across large acreages, while improving overall crop management decisions.
What is AICropCAM, and How Does it Work?
AICropCAM (Figs. 1, 2, & 3) is an advanced edge image processing platform designed to extract plant and canopy features directly from field crops. Its structure includes three key layers:
One of AICropCAM’s biggest advantages is its ability to perform deep learning-based image processing directly in the field. This means it can detect subtle signs of weed emergence or insect damage in real-time, capturing critical information that traditional imaging or simple sensors often miss. Edge computing also significantly reduces the need for high-bandwidth data transmission, a major limitation in rural agricultural areas.
Looking Ahead
Incorporating AICropCAM into leafy green production systems could help growers proactively manage weeds and insect pressures, optimize resource use, reduce chemical interventions (especially important in organic systems), and save time and labor. These technologies offer a glimpse into the future of precision agriculture, where early detection and informed decision-making can significantly boost sustainability and profitability.
Figure 1 (left): AICropCAM-insect detection, Figure 2 (middle): AICropCAM-weed
detection, Figure 3 (right): AICropCAM-installed in the field.
VegIPM Update Vol. 17, Num. 15
July 22, 2026
Results of trap catches below!!
Whitefly: Adult activity remains steady across locations; above average for this time of the year, especially high numbers seen in North Gila Valley. Historically, whitefly numbers peak in July.
Thrips: Adult thrips activity remained low over the last two weeks. About average for this time of the year. Historically, thrips numbers remain low until Sept-Oct.
