Forecasts and Monitoring System based on WRF model ...satelite.cptec.inpe.br › repositorio9 ›...
1
Forecasts and Monitoring System based on WRF model simulations over La Plata Basin Omar. V. Müller and Miguel A. Lovino, CEVARCAM, FICH, UNL – CONICET. Argentina. [email protected] E. Hugo Berbery , CICS-MD, ESSIC, UMD. USA. [email protected] 2nd WCRP Summer School on Climate Model Development: Scale aware parameterization for representing sub-grid scale processes – Cachoeira Paulista - SP, Brazil, 22nd - 31st January 2018 3. Evaluation of rain and temperature A routine forecasts and monitoring system was developed for South America. It is availabe since 2012 on www.cevarcam.unl.edu.ar/en/forecast and www.atmos.umd.edu/~berbery/research/forecasts droughts, wet spells rainy events, hot or cold waves, … 1. Forecasts and Monitoring System Variables: Precipitation Temperature Winds Evapotranspiration Soil Moisture Runoff 2. Applications 6. References Müller, O. V., Lovino, M. E., Berbery, E. H., (2016). Evaluation of WRF Model Forecasts and Their Use for Hydroclimate Monitoring over Southern South America. Weather and Forecsasting, 31, 1001–1017. 7. Acknowledgments This research was carried out with the support of CRN3035 and CRN3095 projects from the Inter-American Institute for Global Change Research (IAI) supported by the US NSF. 142 observations sites to evaluate 2 years of forecasts. First, contingency tables are computed for each forecasted day and each rain gauge Observed YES Observed NO Total Forecasts YES Hits (h) False alarms (fa) Forecasts YES Forecasts NO Misses (m) Correct negatives (cn) Forecasts NO Total Observed YES Observed NO Total Then, a set of skill scores are calculated to evaluate the performance of precipitation forecasts 4. Monitoring 193 observations sites to evaluate daily min, mean and max temperature. The skill scores are correlation coefficients and mean absolute error. 5. Conclusions The system show high prediction performance with 7-day lead time ~ 70% of rain and no-rain days are correctly predicted. Temperature observations and forecasts are highly correlated (r>0.80) The model products demonstrated its ability to reproduce and complements the observed drought in southern Brazil and the above normal precipitation in northeastern Argentina during the spring of 2014. * Upper LPB presented lower rain events than usual (a). It was notably simulated by the model (b), which complements the observations showing the deficit of SM, EVT, and R that reduced the river flows and the provision of water to metropolitan areas. * Lower LPB presents above normal precipitation events (c). The model shows that the wet anomalies was partitioned between EVT, SM, and R (d). This condition favored the planting of corn and soybean in high terrain, but it produced losses over lowlands. Weather forecasting and monitoring systems based on regional models are becoming increasingly relevant for decision support in agriculture and water management, but HOW RELIABLE ARE THESE FORECASTS? Goal: To evaluate the predictive and monitoring capabilities of the real-time forecast system based on numerical simulations with the WRF Model in a region of high agricultural productivity and water needs. Monitoring of wet spell or drought Forecasts of weather events Rain storms Heat waves 15 km Weed growth risk Flow forecasts 45 km Results:
Transcript of Forecasts and Monitoring System based on WRF model ...satelite.cptec.inpe.br › repositorio9 ›...
Forecasts and Monitoring System based on WRF model simulations over La Plata Basin
Omar. V. Müller and Miguel A. Lovino, CEVARCAM, FICH, UNL – CONICET. Argentina. [email protected] E. Hugo Berbery, CICS-MD, ESSIC, UMD. USA. [email protected]
2nd WCRP Summer School on Climate Model Development: Scale aware parameterization for representing sub-grid scale processes – Cachoeira Paulista - SP, Brazil, 22nd - 31st January 2018
3. Evaluation of rain and temperature
A routine forecasts and monitoring system was developed for South America. It is availabe since 2012 on www.cevarcam.unl.edu.ar/en/forecast and www.atmos.umd.edu/~berbery/research/forecasts
droughts, wet spells rainy events, hot or cold waves, …
1. Forecasts and Monitoring System
Variables:
Precipitation
Temperature
Winds
Evapotranspiration
Soil Moisture
Runoff
2. Applications
6. ReferencesMüller, O. V., Lovino, M. E., Berbery, E. H., (2016). Evaluation of WRF Model Forecasts and Their Use for Hydroclimate Monitoring over Southern South America. Weather and Forecsasting, 31, 1001–1017.
7. AcknowledgmentsThis research was carried out with the support of CRN3035 and CRN3095 projects from the Inter-American Institute for Global Change Research (IAI) supported by the US NSF.
142 observations sites to evaluate 2 years of forecasts. First, contingency tables are computed for each forecasted day and each rain gauge
Observed YES Observed NO Total
Forecasts YES Hits (h) False alarms (fa) Forecasts YES
Forecasts NO Misses (m) Correct negatives (cn) Forecasts NO
Total Observed YES Observed NO Total
Then, a set of skill scores are calculated to evaluate the performance of precipitation forecasts
4. Monitoring
193 observations sites to evaluate daily min, mean and max temperature. The skill scores are correlation coefficients and mean absolute error.
5. Conclusions The system show high prediction performance with 7-day lead time ~ 70% of rain and no-rain days are correctly predicted. Temperature observations and forecasts are highly correlated (r>0.80) The model products demonstrated its ability to reproduce and complements the observed drought in southern Brazil and the above normal precipitation in northeastern Argentina during the spring of 2014.
* Upper LPB presented lower rain events than usual (a). It was notably simulated by the model (b), which complements the observations showing the deficit of SM, EVT, and R that reduced the river flows and the provision of water to metropolitan areas.* Lower LPB presents above normal precipitation events (c). The model shows that the wet anomalies was partitioned between EVT, SM, and R (d). This condition favored the planting of corn and soybean in high terrain, but it produced losses over lowlands.
Weather forecasting and monitoring systems based on regional models are becoming increasingly relevant for decision support in agricultureand water management, but HOW RELIABLE ARE THESE FORECASTS?
Goal: To evaluate the predictive and monitoring capabilities of the real-time forecast system based on numerical simulations with the WRF Model
in a region of high agricultural productivity and water needs.
Monitoring of wet spell or drought Forecasts of weather events
Rain storms
Heat waves
15 km
Weed growth risk Flow forecasts
45 km
Results:
