Predictive models of minimum temperatures for the south of Buenos Aires province

HERNANDEZ, G.; MÜLLER, G.V.; VILLACAMPA, Y.; NAVARRO-GONZALEZ, F.J.; ARAGONÉS, L.

THE SCIENCE OF TOTAL ENVIRONMENT

Elsevier B.V

Año: 2019

0048-9697

Depending on the time of development of a crop temperature below 0 ° C can causedamage to the plant, altering its development and subsequent yield. Since frosts areidentified from the minimum air temperature, the objective of this research paper is togenerate forecast -(predictive) models at 1, 3 and 5 days of the minimum dailytemperature (Tmin) for Bahía Blanca city. Non-linear numerical models are generatedusing artificial neural networks and geometric models of finite elements. Sixindependent variables are used: temperature and dew point temperature atmeteorological shelter level, relative humidity, cloudiness observed above the station,wind speed and direction measured at 10 m altitude. Data have been obtainedbetween May and September from 1956 to 2015. Once the available data had beenanalysed, this period was reduced to 2007-2015. For the selection of the most suitablemodel, the correlation coefficient of Pearson (R), the determination coefficient (R2) andthe Mean Absolute Error (MAE) are evaluated. The results of the study determine thatthe geometric model of finite elements with 4 variables, over 9 years (2007-2015) andseparated by the season of the year is the one that presents better adjustment in theforecast of Tmin with up to 5 days of anticipation.