Abstract

AMARIS, Gloria; AVILA, Humberto  and  GUERRERO, Thomas. Applying ARIMA model for annual volume time series of the Magdalena River. Tecnura [online]. 2017, vol.21, n.52, pp.88-101. ISSN 0123-921X.  https://doi.org/10.14483/udistrital.jour.tecnura.2017.2.a07.

Abstract Context: Climate change effects, human interventions, and river characteristics are factors that increase the risk on the population and the water resources. However, negative impacts such as flooding, and river droughts may be previously identified using appropriate numerical tools. Objectives: The annual volume (Millions of m³/year) time series of the Magdalena River was analyzed by an ARIMA model, using the historical time series of the Calamar station (Instituto de Hidrología, Meteorología y Estudios Ambientales de Colombia-IDEAM), and looking for matching the modelling hypothesis with the data measured in the river. Methods: The ARIMA model is considered one of the most implemented approaches in hydrology and studies related to climatic variability because it considers non-stationary information. Results: The maximum volume forecasted of the Magdalena River from 2013 to 2024 oscillates between 289,695 million m³ and 309,847 million m³. The minimum volume forecast for the same period ranges from 179,123 million m³ to 157,764 million m³, with a decreasing trend of 106 million m³ in 100 years. Conclusions: The simulated results obtained with the ARIMA model compared to the observed data showed a fairly good adjustment of the minimum and maximum magnitudes. This allows concluding that it is a good tool for estimating minimum and maximum volumes, even though this model is not capable of simulating the exact behaviour of an annual volume time series.

Keywords : statistical model; autoregressive model; time series.

        · abstract in Spanish     · text in Spanish     · Spanish ( pdf )