Use of Precipitation Duration Data from Weather Radar

Use of Precipitation Duration Data from Weather Radar in Leaf Wetness Duration Estimates for Plant Disease Management MSc. Thesis, University of Guelph Tracy Rowlandson Iowa State University MWWG Presentation

Outline • • Introduction Background Study Objectives Use of radar rainfall estimates in disease management schemes • Radar indication of daily rainfall occurrence • Conclusions and future work MWWG Presentation

Introduction • Decisions regarding fungicide spray timings are often based on leaf wetness duration (LWD) and temperature during that period • < 1 mm of rain will remain on a leaf • Occurrence is more important than quantity • Rain events are spatially diverse so data captured by a rain gauge is site-specific MWWG Presentation

Background • Common disease management schemes include Tom. Cast and Mel. Cast • Require input of LWD and average temperature in order to estimate disease index values • Empirical and physical models have been developed to predict LWD • Physical models based on the concept of energy budgets MWWG Presentation

Leaf Wetness Sensors • Electronic sensors look at changes in electrical resistance • Printed circuit board of gold-plated copper contacts • Sensors are coated with paint and dried at high temperatures MWWG Presentation

Radar Basics MWWG Presentation

Study Objectives • Adapt the Penman-Monteith model to: • Simulate the wetness duration on a tipped leaf surface • Determine the length of the drying period following the end of a rainfall event • Determine if radar is a valuable substitute or complement to tipping bucket rain gauge networks to estimate duration of rainfall events in disease management. MWWG Presentation

Disease Index Estimations Methodology • RH, temperature, windspeed, solar radiation, and longwave radiation (at Elora) were measured and averaged over hourly periods • Solar radiation was adjusted to represent the amount that was received on the tipped surface of the leaf wetness sensor • P-M model was used to indicate the amount of time required for the sensor to dry after a rain event. MWWG Presentation

Disease Index Estimations Methodology • Radar Duration = combination of rainfall duration indicated by the radar, and the drying time indicated by the P-M model • TBRG Duration = combination of rainfall duration indicated by the TBRG, and the drying time indicated by the P-M model MWWG Presentation

Disease Index Estimations Methodology • When rain was indicated by the radar, the rain event was given a 0. 6 mm value in the rain reservoir the hour following the indicated rain • The duration of the wetness period was determined • Mean temperature during the wetness period was calculated • If a break in the wetness period was 2 hours or greater, the wetness event was separated into two events MWWG Presentation

Disease Index Estimations Methodology • If the wetness event was longer than 24 hours, a break in the wetness event was enforced • Disease severity values (DSV’s) and Environmental Favorability Indices (EFI’s) were calculated on a daily basis beginning at 1100 and ending 1100 the following day • If a wetness event extends beyond 1100, the user can extend the period to 1400 (for a maximum wetness duration of 27 hours) MWWG Presentation

Disease Index Estimations Methodology • DSV/EFI accumulations were made using both conventional and Doppler radars • At Elora, estimations were made using all 4 radars • At Ridgetown, range limitations of the King City Doppler radar prevented it from being used • DSV’s/EFI’s were also calculated using rainfall from TBRG at Elora and Ridgetown • Leaf wetness sensor was considered to be the best indicator of LWD and was used as the basis for comparison MWWG Presentation

Radar Basics MWWG Presentation

Disease Index Estimations Results June 9 -11, Elora Time MWWG Presentation

Disease Index Estimations - Results Avg. Temp. Tom. Cast DSV Mel. Cast EFI 19 17. 3 2 4 Exeter conventional 21 17. 5 3 6 Exeter Doppler 21 17. 5 3 6 King City conventional 22 16. 4 3 4 King City Doppler 14 14. 3 1 1 TBRG 7 13. 1 1 1 June 9 Duration (hours) Sensor DSV and EFI calculations at Elora MWWG Presentation

Disease Index Estimations – Results June 11 Duration (hours) Avg. Temp Tom. Cast DSV Mel. Cast EFI Sensor 10 8. 1 0 0 Exeter conventional 5 17. 7 1 1 10 8. 1 0 0 King City conventional 13 9. 9 0 0 King City Doppler 10 8. 1 0 0 TBRG 10 8. 1 0 0 Exeter Doppler MWWG Presentation

Disease Index Estimations - Results June 10 Duration (hours) Avg. Temp Tom. Cast DSV Mel. Cast EFI Sensor 11 17. 3 2 4 Exeter conventional 24 17. 5 3 6 Exeter Doppler 24 17. 5 3 6 King City conventional 13 16. 4 3 4 King City Doppler 12 14. 3 1 1 3 14. 8 0 0 12 10. 5 0 0 TBRG MWWG Presentation

Disease Index Estimations Results Sensor TBRG Exeter conv. Exeter Dop. King City conv. King City Dop. DSV EFI DSV EFI 27 45 27 51 41 67 40 66 28 54 25 41 Cumulative Error 0 +6 +14 +22 +13 +21 +1 +9 -2 -4 Total DSV and EFI calculations for Elora MWWG Presentation

Disease Index Estimations Results Sensor TBRG Exeter conv. Exeter Dop. King City conv. DSV EFI DSV EFI 39 69 49 90 53 101 56 112 54 104 Cumulative Error +10 +21 +14 +32 +17 +43 +15 +35 Total DSV and EFI calculations for Ridgetown MWWG Presentation

Disease Index Estimations Conclusions • Errors due to excessively long wet periods • Radar indicates rainfall for too long • Overestimation of the length of drying time when radar indicates sporadic rainfall • Rating of the radars • At Elora, the King City radars were the most capable of mimicking the sensor • At Ridgetown, the Exteter conventional and King City conventional radars faired the best MWWG Presentation

Study Conclusions • Study objective: Determine if radar is a valuable substitute or complement to tipping bucket rain gauge networks in plant disease management schemes • Radar rainfall estimates should be used as a complement • The allowable window for error in Mel. Cast is too small for the use of radar rainfall estimates • TBRG and P-M model mimicked sensor quite well • Use of radar for as a check for interpolation should be done with caution MWWG Presentation

Future Research Advancements • User could establish an automated system that tests various threshold values for current weather conditions or daily estimate. • Radar rainfall estimates could be used to determine if a TBRG is installed correctly, need calibration or maintenance or has been tampered with. • Further investigation into the use of radar rainfall estimates in the timing of fungicides sprays. MWWG Presentation

Questions? MWWG Presentation