New York is the most populous city in the United States, the second largest in North America and is among the five largest megacities in the world. However has a relatively harsh climate with cold winters and hot, humid summers. In recent years, records show an increase in extreme events such as heat waves, polar waves, above average rainfall and hurricanes. That is why an accurate prognosis, it is essential to reduce uncertainties to mitigate the risk of this population.
Figure 1. Weather data for the city of New York (Annual)
Figure 2. Summary of all time
Figure 3. Monthly Summary
Figure 4. Annual extreme temperatures (February to December) 1910-2013
Figure 5. Extreme annual rainfall (February to December) 1910-2013
Figure 6. Prediction of maximum temperatures
Figure 7. Prediction of minimum temperatures
In the forecasts made with methods of Artificial Intelligence for the city of New York, particularly using a combination of Genetic Programming and Artificial Neural Networks, excellent results which exceed current forecasters were obtained. The results have greater than 90% of certainty in every case tested for minimum temperature and 80% for maximum ones. higher temperatures, having reached optimal solutions with less than 5% errors.
These results demonstrate the scope and power these tools have to save uncertainties in daily conflicts related to meteorological variables, among others, where it is possible to make very accurate predictions based on historical data. In BerecoLabs we are developing these approaches and verifying a significant potential to understand and manage complex systems of reality.