Fire blight prediction models — words and colors (and numbers?) matter

This week apples have been in bloom in Massachusetts. Along comes the risk of fire blight blossom blight. Warm temperatures and moisture are known to be the trigger for blossom infection by fire blight. It’s been relatively cool — 60’s to lower 70’s — so the apparent risk of infection has been perceived as low. In addition, we have fire blight models on NEWA (Network for Environment and Weather Applications). And they — both Cougar Blight and the EIP (Epiphytic Infection Potential based on the MaryblytTM model) were showing generally low risk for blossom infection by fire blight. Nice, breathe easy, right?

But wait, there is always a catch when it comes to dealing with fire blight. A grower, who has a history of fire blight in their orchard, called me in the afternoon and said something to this effect, “Jeez Jon, this morning NEWA was showing low risk of fire blight infection but now one of them is green (Cougar Blight) and the other (EIP, Maryblyt) is showing HIGH, ORANGE risk of infection. What gives? Do I need to spray streptomycin? Arghhh, I had just put a cover spray on for scab this morning but did not include strep and I really don’t want to have to go out and spray the whole orchard again to prevent fire blight, what should I do?”

Great. Begs many questions. For one, I told the grower I am never going to tell him definitively NOT to spray strep for fire blight! But seriously, the day turned out warmer (low 70’s) than predicted (mid 60’s) so the model output changed. But why the perceived difference (by the grower) in model outputs, Cougar Blight being GREEN (low risk of infection?) while the EIP was ORANGE signifying much greater risk of infection? I was unable to give them a clear answer, so I had to do a little digging to refresh my memory on how the models differ in estimating risk of infection. I want to say here, however, the user-interface (on NEWA) is not the best and I hope my discourse might result in some clarification of the fire blight model outputs on NEWA going forward? Kerik?

OK, all a grower needs to know about the two fire blight models on NEWA, Cougar Blight and EIP.

Cougar Blight simply accumulates degree hours as heat units beginning at bloom. Then, models the fire blight infection risk level based solely on these accumulated heat units/degree hours. Note that it does NOT use wetting or anything else in determining risk, you have to assess the level of wetting's contribution to fire blight infection, be it rain, dew, spray event, etc. But if Cougar Blight says the risk is HIGH or particularly EXTREME you better be applying some streptomycin to open bloom if wetting occurs. Note that the infection risk level also changes based on the amount of fire blight recent history in the neighborhood. Generally I leave it as fire blight occured in the neighborhood last year to keep it middle of the road unless I know fire blight is currently active. Very simple, but realize the Cougar Blight model can say HIGH or EXTREME but if there is no wetting event, there will be no infection. Here is the latest on Cougar Blight from WSU, but realize they talk about wetting event in their DAS which is not included in the NEWA Cougar Blight model. 

EIP as I said is straight out of the MaryBlyt model. Which is way more complicated than Cougar Blight! (Which makes it better? Maybe?) I will try to boil it down to as simple as possible. For a blossom blight infection to actually occur, regardless of whether EIP/Maryblyt shows up YELLOW or ORANGE, four condidions must be met:

1. open bloom (D'oh! as Bart would say)
2. degree hours/heat units accumulation threshold
3. average (daily, 24 hours) temperature greater than or equal to 60 degrees F.
4. AND wetting! one tenth of an inch or more of rain, or could be heavy dew or dilute spray event?

INFECTION will not occur unless all four of these conditsion are met. Just one, two, or three of these conditions might indicate LOW, MODERATE, or HIGH risk of infection, but you would still need the fourth parameter, likely wetting, for INFECTION to actually occur and eventually show up as blossom blight. Below is a Table -- straight out of the 44 page Maryblyt manual, available here -- that pretty much sums it up:

Ok, let's look at the NEWA fire blight model ouptut that triggered this blog post:

The afternoon in question was 5/3, this model output was run after the fact, and actually at the time on 5/3 Cougar Blight was showing a GREEN table cell. You can see why the grower might have been concerned, not understanding the nuances of the models, a GREEN-Low infection risk cell table in Cougar Blight, and an ORANGE-High infection risk in EIP (Maryblyt). Huh? Admittedly, NEWA does, below the Table provide this explanation: 

Somewhat helpful but not overly, especially if you are panicking trying to figure out if I should apply a strep spray or "Oh crap, how come I did not include strep in that scab spray I just put on this morning???" Really, I told the grower it was in fact borderline and forecast cooler temperatures might reduce the risk of infection significantly. They still went out and sprayed streptomycin on younger, more susceptible blocks, or maybe it's been rumored they sprayed the whole orchard! Was there actually an INFECTION? Probably not, no, it appears not all four conditions were met for an INFECTION to occur. Bloom? Check. Degree hours? Check, barely, briefly, but then went down the next day 5/4. Wetting. Check, but not until next day 5/4. Average Temperature? Check, but only for 5/3 and barely. NO INFECTION! No need to spray strep! If you have faith and believe the models are right?

My thought is growers really need to know that all four conditions -- bloom, degree hour accumulation, average temperature, and wetting need to be met for an INFECTION to actually occur. This again in the Maryblyt manual is helpful: "A blossom [strep] spray is recommended when the risk is HIGH and when an infection is predicted for the next day."

Oh, and then there is RIMpro...

Whew, not sure I really cleared anything up or simplified it or not? Note to self, stick with horticulture, want to talk about thinning sprays now??? :-)