Technical Documentation and Links for Plant Disease Risk and Other Hourly Weather Driven Models

Len Coop
Assoc. Dir. Decision Support Systems
Oregon IPM Center
Oregon State University

Our disease model system accesses hourly and sub-hourly data from numerous real-time and archival weather stations (>33,000 stations, 2021) to compute a variety of plant disease risk and horticultural models. Currently the following models are available:

1. Apple Scab infection risk
2. Pear Scab infection risk
3. Gubler Thomas Powdery Mildew
4. Hop Powdery Mildew
5. Cherry Powdery Mildew
6. Pearson-Gadoury Powdery Mildew
7. Cleistothecial Powdery Mildew
8. Strawberry Powdery Mildew
9. Anjou Pear Scald
10. Fire Blight (Cougar Blight)
11. Fox-Broome Botrytis infection risk
12. Blueberry Mummy Berry
13. Boxwood Blight infection risk
14. Dollar Spot (turfgrass)
15. Potato / Tomato Late Blight
16. TomCast Disease Severity Values (DSV)
17. MelCast Environmental Favorability Index (EFI)
18. Vapor Drift Risk Estimation Model
19. Custom Degree-Hour Accumulation Model
20. Chilling Hours models

Use these models with caution. Please see our Disclaimer.

• Apple Scab Infection Risk

  • This disease, caused by the fungus Venturia inaequalis, is documented in the PNW Plant Disease Management Handbook.
  • Models infection risk only. Requires hourly data.
  • Model is an inverted Mills Table; developed as a degree-hour version of the Stensvand modification of the standard Mills table for apple scab infection risk. See Stensvand A, Gadoury DM, Amundsen T, Semb L, Seem RC, 1997. Ascospore release and infection of apple leaves by conidia and ascospores of Venturia inaequalis at low temperatures. Phytopathology 87, 1046–53.
  • This version by Len Coop, documented at OSU IPPC.
  • The algorithm used is:

            if ( $lfwetness < 1 ) {
                              $Anomoistcount++;
                              $ADH = 0;
                              if ( $Anomoistcount > 8.8 ) { $AcumDH = 0 }
              }
              else {
              $Anomoistcount = 0;
              if ( $temp < 32 ) { $ADH = 0 }
                           else {
                           if ( $temp > 66 ) { $ADH = 66 - 30 }
                else { $ADH = $temp - 30 }
                $AcumDH += $ADH;
                }
                }
                if ( $AcumDH < 175 )           { $Alabel = $gre . "no app_scab" . $stp }
                               if ( $AcumDH >= 175 )          { $Alabel = $yel . "scab near  " . $stp }
                  if ( $AcumDH > 204 )           { $Alabel = $red . "APPLE SCAB!" . $stp }
                  if ( $AcumDH > 275 )           { $Alabel = $red . "SCAB cycle!" . $stp }
          

  • See also
  • This model is available via our web app and our email subscription service.

• Pear Scab Infection Risk

  • This fungal disease, caused by Venturia pirina, is documented in the PNW Plant Disease Management Handbook.
  • Models infection risk only. Requires hourly weather data.
  • Developed by Spotts and Cervantes; see Spotts, R. A. and Cervantes, L. A. 1991. Effects of Temperature and Wetness on Infection of Pear by Venturia pirina and the Relationship Between Preharvest Inoculation and Storage Scab. Plant Disease 75:1204-1207. Converted for degree-hour calculations by Coop and Spotts (2002).
  • This version is documented at OSU IPPC.
  • The algorithm used is:

            if ( $lfwetness < 1 ) {
                              $Pnomoistcount++;
                              $PDH = 0;
                              if ( $Pnomoistcount > 11.8 ) { $PcumDH = 0 }
              }
              else {
              $Pnomoistcount = 0;
              if ( $temp < 32 ) { $PDH = 0 }
                           else {
                           if ( $temp > 66 ) { $PDH = 66 - 32 }
                else { $PDH = $temp - 32 }
                $PcumDH += $PDH;
                }
                }
                if ( $PcumDH < 250 )           { $Plabel = $gre . "no pearscab" . $stp }
                               if ( $PcumDH >= 250 )          { $Plabel = $yel . "scab near  " . $stp }
                  if ( $PcumDH > 320 )           { $Plabel = $red . "PEAR SCAB! " . $stp }
                                        if ( $PcumDH > 350 )           { $Plabel = $red . "SCAB cycle!" . $stp }
    

  • There is a related degree-day "end of primary ascospore season" model for pear scab, also at uspest.org.
  • See also http://www.ipm.ucdavis.edu/DISEASE/DATABASE/pearscab.html for more model documentation.
  • This model is available via our web app and our email subscription service.

• Gubler Thomas powdery mildew risk index

  • This disease, caused by the fungus Erysiphe necator (Also called Uncinula necator), is documented for grapes in the PNW Plant Disease Management Handbook and by the UC Statewide IPM Program.
  • Well known index model for powdery mildew of many crops including grapes. Models conidial stage infection risk only. Requires hourly weather data.
  • This model is published at Gubler et al. 1999 - APSnet Online
  • This model is available via our web app and our email subscription service.

• Hop Powdery Mildew

  • This disease, caused by the fungus Podosphaera macularis, is documented in the PNW Plant Disease Management Handbook.
  • This is actually two models, an older model for highly susceptible varieties, and a newer model for "Cascade" and similar varieties. Both are models of risk for the conidial stage only. Requires hourly weather data. Developed for use in the Pacific Northwest.
  • The older model was developed by Mahaffee and Gent, modified from the Gubler-Thomas powdery mildew model. An early version of the Hop PM model is documented at Mahaffee, W. F., Thomas, C. S., Turechek, W. W., Ocamb, C. M., Nelson, M. E., Fox, A. Gubler, W. D. 2003. Responding to an introduced pathogen: Podosphaera macularis (hop powdery mildew) in the Pacific Northwest. Online. Plant Health Progress doi:10.1094/PHP-2003-1113-07-RV.
  • The Cascade model, developed by Dave Gent, has not been published as of Feb 2020. It is similar to the older model, but with these rules:
    1. If there were ≥6 continuous hours at ≥28°C, then subtract 20 points, else;
    2. If there were ≥2 continuous hours at ≤4°C, then subtract 10 points; else
    3. If there were ≥2.5 mm rain, then subtract 10 points, else;
    4. If there were ≥6 continuous hours at ≥28°C on the previous day, then there is no change in the index, else;
    5. If there were ≥6 continuous hours of temperatures from 16 to 27°C, then add 20 points, else;
    6. If none of the above rules apply, then subtract 10 points.
    Index values still accumulated over time, with minimum and maximum values of 0 and 100, respectively. Values of 0 to 30, 40 to 60, and 70 to 100 indicate conditions of low, moderate, or high infection risks, respectively.
  • These models are available via our web app and our email subscription service.

• Cherry Powdery Mildew

  • This disease, caused by the fungus Podosphaera clandestina, is documented in the PNW Plant Disease Management Handbook and WSU.
  • Grove model of Powdery Mildew Model infection risk. Models conidial stage only.
  • The basic Cherry PM model has not been fully documented and at this time is considered to be experimental. Use at your own risk. Requires hourly weather data.
  • Developed by G. Grove et al for Washington Sweet Cherries. See Grove, Gary G., and R. J. Boal. 1991b Powdery mildew of sweet cherry: Influence of temperature and wetness duration on release and germination of ascospores of Podosphaera clandestina Phytopathology 81.10: 1271-1275.
  • This model is available via our web app and our email subscription service.

• Pearson-Gadoury (1987) Powdery Mildew Model

  • Grape powdery mildew, caused by Erysiphe necator (Also called Uncinula necator), is documented in the PNW Plant Disease Management Handbook.
  • Model of primary powdery mildew infection in grapes; alternative to Cleistothecial PM model. The model is used during bud break and early shoot growth.
  • Requires hourly weather data. It was developed in New York State and has been adopted for use in Western Oregon.
  • This model is very simple: ascospores are released from overwintered cleistothecia (or chasmothecia) on days when the average daily temperature exceeds 10°C (50°F.) AND daily rainfall (or other form of precipitation) exceeds 0.1 inches (2.5 mm).
  • References include Cornell Extension papers: Gadoury et al. (2012) and Gadoury et al. (2006).

• Cleistothecial Powdery Mildew

  • Grape powdery mildew, caused by Erysiphe necator (Also called Uncinula necator), is documented in the PNW Plant Disease Management Handbook.
  • This model is for the primary inoculation of grape by powdery mildew ascospores released from cleistothecia, and is often used in conjunction with the Gubler-Thomas Powdery Mildew model, expecially in California.
  • Risk index model for initial infection in grape by ascospores, also known as cleistothcia. Requires hourly weather data.
  • Model source: Thomas, C. S., Gubler, W. D., and Leavitt, G. 1994. Field testing of a powdery mildew disease forecast model on grapes in California. Phytopathology 84:1070 (abstr.). This model is a version of the model also known as the "2/3 Mills table" model.
  • Also documented by the UC Statewide IPM Program.

• Strawberry Powdery Mildew Index Model

  • This disease, caused by Podosphaera aphanis var. aphanis (formerly Sphaerotheca macularis f. sp. fragariae) is documented in the PNW Plant Disease Management Handbook.
  • This is a risk index model, and requires hourly weather data.
  • Model is similar to Gubler-Thomas Powdery Mildew with thresholds determined by Miller et al. See Miller, T. C., Gubler, W. D., Geng, S., and Rizzo, D. M. 2003. Effects of temperature and water vapor pressure on conidial germination and lesion expansion of Sphaerotheca macularis f. sp. fragariae. Plant Dis. 87:484-492.

• Pear Scald - Anjou Pear Model

  • This storage problem is documented in the PNW Plant Disease Management Handbook for apples and for pears
  • Developed to estimate safe storage time based on temperatures prior to harvest. Requires hourly weather data.
  • This model was developed by Jinhe Bai at OSU, and is documented at OSU IPPC.

• Fire Blight (Cougar blight)

  • Disease of pome fruits caused by Erwinia amylovora. Documented at WSU tree fruit site, in the PNW Plant Disease Management Handbook, and by WSU Extension.
  • Two different models for fire blight (cougar blight) infection risk. Cougarblight 2010 is the preferred model, which requires hourly weather data. Cougarblight 2010EZ is a simplified model which only requires daily maximum and minimum temperatures. The full 2010 model is preferred.
  • These model are part of an ongoing series developed by Tim Smith at WSU, and are documented in Smith, T.J., and P. L. Pusey. 2010 CougarBlight 2010 ver. 5.0: a significant update of the CougarBlight fire blight infection risk mode. Acta Horticulturae 896:331-338.
  • This model is available via our web app and our email subscription service.

• Fox-Broome Botrytis infection risk

  • Diseases are caused by Botrytis cinerea include grape bunch rot (bunch mold), documented in the PNW Plant Disease Management Handbook, and many others in other crops.
  • Used regularly for grape bunch mold and for Botrytis cinerea diseases in other fruit crops. Requires hourly weather data.
  • The model source is Broome, J. C., English, J. T. Marois, J. J., Latorre, B. A. and Aviles, J. C. 1995. Development of an Infection Model for Botrytis Bunch Rot of Grapes Based on Wetness Duration and Temperature. Phytopathology 85: 97-102. It is also described on the UC Statewide IPM site (model 1 of 2).
  • This model is available via our web app and our email subscription service.

• Blueberry Mummy Berry

  • This disease is documented in the PNW Plant Disease Management Handbook.
  • This is a new/preliminary degree-hour/leaf wetness model to predict ascospore infection risk in blueberry. It requires hourly weather data.
  • The model is based on work by Hildebrand and Braun 1991 Can. J. Plant Path. 13:232-240, see Fig. 5) and is documented in this spreadsheet.
  • There is a related degree-day model of the "end of primary ascospore season", also at uspest.org and also documented in the spreadsheet linked above.

• Boxwood Blight Infection Risk

  • This disease is caused by Calonectria pseudonaviculata, and is documented in the PNW Plant Disease Management Handbook, Purdue Extension, and Boxwood Blight Knowledge Center.
  • This model estimates first infections and degree of infection risk. This is hourly version 3.0; it requires hourly weather data.
  • Limitations: As of summer 2023, this model is thought to fairly reflect infection potential for the two susceptible varieties it was based upon. Like all infection risk models, we assume that infectious inoculum is present. The model currently assumes that infection occurs during periods of leaf wetness caused by dew or rain. In some regions such as the southeastern US, high humidity may contribute to leaf wetness not currently accounted for in this model. The model may therefore sometimes underpredict disease under these conditions (high RH).
  • Based on work by Gehesquière (2014), Avenot et al. (2017, 2021), and Weiland et al. (2022). A technical summary of the model at uspest.org is at: model version 3 summary document (pdf). A guide to using the model is at: model usage instructions (pdf).
  • This model is available via our web app and our email subscription service.

• Dollar Spot (turfgrass)

  • Dollar Spot Disease, caused by fungal pathogen Sclerotinia homoeocarpa, is described at UC Davis Dollar Spot, Purdue University Dollar Spot, and APS Dollar Spot of turfgrass.
  • The dollar spot model predicts risk of dollar spot epidemics in turfgrass. Thresholds for treatment are likely to need local calibration. Currently the authors are recommending a threshold of 20% probability for an epidemic. The model includes a parameter for fungicide treatment within the last 15 days, which will lower the probability of an epidemic.
  • The geographic range of the model may be limited to those areas thus far tested (Wisconsin, Oklahoma, Pennsylvania, Tennessee, and Mississippi).
  • Like most weather driven risk models, this one assumes that sufficient inoculum is present for causing infections. It uses 5-day average temperature and relative humidity to predict the relative probability of an epidemic, and need for management using fungicides, so it requires hourly weather data.
  • This model is from work by Smith, Kerns, et al, published as Smith DL, Kerns JP, Walker NR, Payne AF, Horvath B, Inguagiato JC, et al. (2018) Development and validation of a weather-based warning system to advise fungicide applications to control dollar spot on turfgrass. PLoS ONE 13(3): e0194216. https://doi.org/10.1371/journal.pone.0194216. See also this model description and validation summary (pdf).
  • See also U. Wisc. Dollar Spot Model, which describes the model and its use.
  • This model is available via our web app and is available through our email subscription service.

• Potato / Tomato Late Blight

  • Late blight is caused by the fungus Phytophthora infestans in potato and tomato. The disease is documented for tomato and for potato in PNW Plant Disease Management Handbook.
  • Requires hourly weather data. Model may underestimate disease when irrigation increases leaf wetness or humidity.
  • This model based on the Smith and the Winstel versions, and is calculated at 6 am each day from prior 24 hr data. It uses these rules:
    1. If 10 consecutive hours of (leafwetness > 3 OR relative humidity > 90 % ) AND temperature > 10°C, then there is an "infectious spore production" event.
    2. If "infectious spore production" event is at least 24 hours old and no more than ten days old, and during that time there are two consecutive days with a maximum temperature between 23 and 30°C, there is an "infection possible" event is.
    3. The "infections possible" is active for three days. Lesions are likely to be visible after that.
    4. Otherwise the late bight risk is "low".
  • See http://www.ipm.ucdavis.edu/DISEASE/DATABASE/potatolateblight.html. See Model 4 for Winstel version: Winstel, K. 1993. Krautund knollenfaule der kartoffel eine eeue prognosemoglichkeit-sowie bekampfungsstrategien. Med. Fac. Landbouww. Univ. Gent, 58/3b, and See Model 14 for Smith version: Smith, L. P. 1956. Potato blight forecasting by 90% humidity criteria. Plant Pathology 5:83-87.
  • This model is available via our web app and our email subscription service.

• TomCast Disease Severity Values (DSV)

  • This is a widely-used model for several diseases including:
    • Alternaria solani (black mold) on tomato
    • Alternaria leaf blight on carrot
    • Septoria apiicola on celery
  • Model is used to estimate units of disease development. Requires hourly weather data.
  • Disease Severity Values are available at: UC Statewide IPM site.
  • Model source: Pitblado, R. E. 1992. The development and implementation of TOM-CAST: A weather-timed fungicide spray program for field tomatoes. Ministry of Agriculture and Food, Ontario, Canada.

• MelCast Environmental Favorability Index (EFI)

  • There are two versions of this model, for muskmelon and for watermelon. It is used to help with decisions to treat several diseases, including Alternaria leaf blight, gummy stem blight, and anthracnose.
    
  • Extension references include:
    • Purdue Extension Bulletin
    • U. Missouri Extension Webpage
    • U. Delaware Extension Webpage
  • Model is used to help time fungicide sprays. Requires hourly weather data.
  • Model source: Everts, K. L., R. C. Korir, M. J. Newark 2016. Re-evaluation of MelCast for Fungicide Scheduling in Mid- Atlantic Watermelon. APS Plant Health Brief
  • This model is available via our Muskmelon MelCast web app and our Watermelon MelCast web app and our email subscription service.

• Delta-T Vapor/Thermal Drift Risk

  • A tool to help with spray drift avoidance/mitigation. Requires hourly weather data.
  • This is a model adapted from a tool used in Australia to reduce/avoid vapor or thermal drift by monitoring Delta-T (the difference between wet and dry bulb Temperatures; the greater the difference, the more likely small spray droplets will volatilize (vaporize) into the atmosphere and thus become an invisible form of spray drift.
  • This model can be especially useful in arid climates such as the Western US during summer months. It may also be useful to maximize irrigation efficiency (avoid irrigating during periods of high Delta-T).
  • Reference at: Weather for Pesticide Spraying.pdf

• Custom Degree-Hour Accumulation Model

  • A tool for plant disease epidemiologists to diagnose new and invasive plant diseases, and to develop new models
  • This tool can be configured to accumulate degree-hours above a lower threshold and below an upper threshold, as specified by the user.
  • The option of whether these accumulations depend upon leaf wetness is also available. With leaf wetness required, the model can emulate other models such as apple scab, pear scab, cleistothecial powdery mildew, and TomCast DSV.
  • Without the leaf wetness requirement, this tool can be used for such purposes as chilling requirements and general degree-hour accumulations.
  • Use trial and error to configure new model parameters for diseases for which you have field data but no published model. This allows this website to be used for invasive species when models have been hypothesized but not yet researched or validated.

• Chilling Hours models used in Horticulture to predict completion of dormancy

  • Chilling requirements (hours or units) are typically computed to estimate when fruit trees complete dormancy. Insufficient chill hours can result in delayed foliation, and reduced fruit set and quality.
  • This calculator allows both the use of a simple count of hours between two temperature thresholds, such as above 32F and below 45F, and the more complex Utah model (Richardson, Seeley, Walker 1977).
  • See UC Extension for more complete introductions to chill units and formulas for simple and Utah types of chill units.

Last updated 7/7/2023