thank you Dave for the time and effort you put into this post - short of input from specmeters, I'll slightly exaggerate and say that the Cornell model will predict scab infection after a short heavy dew, while the WA model requires rainforest conditions for a couple of days before it indicates concern - I made the assumption that if growing McIntosh in NY one had better be very careful, while desert conditions don't provide the humidity or inoculum to result in scab infections except under the most severe circumstances - I talk to growers at the Ohio River and south and they don't worry much about scab, it's a minor consideration in their climate -
the Mills table (and modified Mills - I attended a number of the MI IPM schools in the '80s when Jones and Howitt and Bird and crew was cutting edge) has been reliable here and referred to for as long as I have been spraying apple trees - I used to get up in the middle of the night, check temperatures, shake limbs to judge free moisture, then refer to the chart to find out just how much trouble I was in - much, much easier to let the instrument keep track of conditions - and until this little event, I hadn't really considered that it would possible to confuse the program - I feel lucky to have some decades of experience to guide decisions, hence the long day today putting a protectant cover on in anticipation of the 4 wet days forecast for the immediate future - I have heard talk from various sources on the ascospore/light association, but have never read any source material - can you point me toward the research that led to this conclusion? I find nothing to disagree with and little to add to your observations about decision factors - thank you - David the Mills table has been an effective tool here On Apr 21, 2011, at 11:13 AM, Dave Rosenberger wrote: > Hello, David -- > I really don't know how the Spectrum instruments are programmed, and I > don't know what they are using for the Washington scab model. Thus, this > response may be of limited usefulness. > I suspect that that the differences between the Cornell and the Mills > (MI) models relate to two key differences among these models. Again, I don't > know this for a fact because I don't know how Spectrum programmed their unit, > but I would guess that the Mills (MI) model uses the modified Mills Table > that was published by Al Jones and he put into the original Reuter-Stokes > scab caster. If so, that program will not discount wetting periods that > start after dark, and it will use slightly longer wetting requirements for > light infections than those used by the Cornell model. > The Cornell model presumably does not count wetting periods that start > after dark because relatively few ascospores are released in the absence of > light. In a high inoculum block, those few that do discharge at night can > still create problems, but in most commercial blocks night-time discharge > will not be significant, especially early in the season. In my opinion, > ignoring night time discharge is a bit more risky by the time one gets to > pink bud (when most ascospores are maturing), and it is absolutely foolhardy > to ignore night time discharges between pink and petal fall if the rains in > question come at the end of a relatively long warm dry period. In this > latter situation, the swelling ascospores cause the pseudothecia to explode > and the light-triggered release mechanism may be by-passed. > The Cornell model was based on lab trials that showed ascospores can > infect in the same relatively short periods that Bill Mills had initially > described for secondary infections. Thus, the Cornell model will trigger > 'infected' before the Mill's MI model. Although NY has adopted the 'Cornell > model' because it is technically more correct, I still prefer the modified > Mill's table that was developed by Jones (although I do believe in > discounting night-time wetting, especially early in the season). > Bill Mills developed his scab model by actually looking at what > happened to trees outdoors. As a result, his model and the modified Mill's > table from Al Jones actually represent an integration of minimum infection > conditions AND spore numbers. Conidia are always produced in much greater > quantities than ascospores under conditions in commercial orchards. Thus, > with conidia, large quantities arrive and infect leaves within 6 hr at > optimum temperatures. If you artificially put large quantities of ascospores > on leaves, you also get infections within 6 hr at optimum temperatures. In > reality, however, it takes some time for an economically significant dose of > ascospores to arrive on leaves in a commercial orchard because there are > relatively few of them. Mills and Al Jones therefore used 9 hr as the minimal > wetting period at optimum temperatures to account for the fact that their > data suggested it would require an extra three hours to accumulate an > economically significant dose of ascospore as compared to conidia. (Some of my plant pathology colleagues my wish to quibble with these broad generalizations because I've skipped a lot of details and also done some "reading between the lines." Nevertheless, I think my general conclusions in comparing the two models are valid.) > The Cornell model also provides only a yes/no response to infection, > whereas the original Mills table and the Jones version of the Mills table > still provides gradations of light, moderate, or heavy infection based on > duration of the wetting periods at various temperatures. Again, because > ascospores are relatively limited in number in most orchards, it makes sense > to parse out the severity of infection for ascospores whereas just the > minimal wetting/temperature requirements are enough to trigger conidial > infections if conidia are present because conidia are either present in large > numbers or not at all. > By having information on light-moderate-heavy infection, one can adjust > one's on-site risk factors based on other details of the specific orchard > situation. For example, in a clean orchard with trees just at green tip, I > would ignore the "light" Mills period and begin to worry only after > triggering at least a "moderate" Mill's period because there are so few > spores at green tip that the marginal conditions for a light infection period > will be unlikely to result in noticeable scab. However, even in a clean > orchard, I would NOT ignore a light Mills period when trees are at tight > cluster or pink. > I hope others will chime in on what differences may be incorporated > into the Washington model. > >> Is anyone else on this list using Spectrum instruments to monitor weather >> and model disease? - I've had a 'Watchdog' for several years now, and >> yesterday had an anomaly, with complete disagreement and inconsistencies >> between the three scab models - >> >> The software uses models from Cornell, Washington State, and Mills (MI) - >> I'm used to Cornell being very conservative and WA being the opposite - this >> event, Cornell indicated 'infected', WA 'none', and Mills 'heavy' - I've >> never modeled an event with Mills being 'heavy' and WA 'none' - >> >> circumstances were such that we were cruising along in the lower 40'sF >> monday and then an overnight rain till 7AM tues morning - a two hour dry >> period, followed by a rain, followed by the violent front (no damage here, >> but tremendous light show) - while that front was moving thru, temperatures >> rose above 50*F for about 6 hours, peaking at 56*, before declining back to >> the lower 40's again - >> >> to add to the mystery, if I model tuesday, from the two hour dry thru the >> end of the event, I get the 'infected', 'none', 'heavy' analysis from the >> program - if I run the model from the start of the rainy event monday so to >> include the rainy monday night/early tues thru the end of the event >> wednesday morning, the Mills model indicates 'light' infection (Cornell >> indicates 'infected', WA 'none) - >> >> as a practical matter, at our stage of development, these temperatures, and >> specific schedule of the wet periods, I normally wouldn't worry much about >> scab infection - but seeing that 'heavy' infection indicated from the Mills >> model is disconcerting - >> >> dunno - any thoughts? >> >> thanks, >> David Doud >> grower, IN >> _______________________________________________ >> apple-crop mailing list >> apple-crop@virtualorchard.net >> http://virtualorchard.net/mailman/listinfo/apple-crop > > > -- > ************************************************************** Dave > Rosenberger > Professor of Plant Pathology Office: 845-691-7231 > Cornell University's Hudson Valley Lab Fax: 845-691-2719 > P.O. Box 727, Highland, NY 12528 Cell: 845-594-3060 > http://www.nysaes.cornell.edu/pp/faculty/rosenberger/ > > _______________________________________________ > apple-crop mailing list > apple-crop@virtualorchard.net > http://virtualorchard.net/mailman/listinfo/apple-crop _______________________________________________ apple-crop mailing list apple-crop@virtualorchard.net http://virtualorchard.net/mailman/listinfo/apple-crop
