Temporal Analysis and Forecast of Surface Air Temperature: case study in Colombia

Document Type : Original Research Paper


1 Facultad de Ingeniería, Universidad Cesmag, Pasto, Colombia

2 Fundaci´ón Universitaria los Libertadores, Bogotá, Colombia


In this work, we study  the short-term dynamics of the Surface  Air Temperature (SAT) using data obtained  from a  meteorological station in Bogotá from 2009 to 2019  and using  time series.  The data that we used correspond to the  monthly mean of the historical registers of SAT and three  pollutants. A descriptive analysis of  the data follows. Then, some predictions are obtained from two different approaches:  (i) a univariate analysis of SAT through a  SARIMA model, which shows a good fit; and  (ii) a multivariate analysis of SAT and  pollutants  using a SVAR model. Suitable transformations were first applied on the original dataset to work with stationary time series. Subsequently, A SARIMA model and a VAR(2) with its associated SVAR model are estimated. Furthermore, we obtain one-year forecasts for the logarithm of SAT in both models. Our forecasts simulate the natural fluctuation of SAT, presenting peaks and valleys in months when SAT is high and low, respectively. The SVAR model allows us to identify certain shocks that affect the instant relationships among variables. These relations were studied by the impulse-response function and the VAR model variance decomposition. Although the statistical methods used in this study are classical, they continue being widely used in the environmental field, presenting god fits, and the results obtained in this study are consistent with  environmental theories.


Agbazo, M., Koto N’gobi, G., Alamou, E., Kounouhewa, B., Afouda, A. and Kounkonnou, N. (2019). Multifractal behaviors of daily temperature time series observed over benin synoptic stations (west africa). Earth Sciences Research Journal, 23(4):365–370.
Aghelpour, P., Mohammadi, B. and Biazar, S. M. (2019). Long- term monthly average temperature forecasting in some climate types of iran, using the models sarima, SVR, and SVR-FA. Theoretical and Applied Climatology, 138(3-4):1471–1480.
Alonso, L. and Renard, F. (2019). Integrating satellite–derived data as spatial predictors in multiple regression models to enhance the knowledge of air temperature patterns. Urban Science, 3(4):101.
Benali, A., Carvalho, A., Nunes, J., Carvalhais, N. and Santos, A. (2012). Estimating air surface temperature in portugal using modis lst data. Remote Sensing of Environment, 124:108–121.
Box, G. E. and Jenkins, G. M. (1976). Time series analysis: forecasting and control revised. Holden–Day.
Grivas, G., Chaloulakou, A. and Kassomenos, P. (2008). An overview of the pm10 pollution problem, in the metropolitan area of athens, greece: assessment of controlling factors and potential impact of long range transport. Science of the Total Environment, 389(1):165–177.
Lozano, N. (2004). Air pollution in Bogotá, colombia: a concentration– response approach. Revista Desarrollo y Sociedad, (54):133–177.
Ninyerola, M., Pons, X. and Roure, J. M. (2000). A methodological approach of climatological modelling of air temperature and precipitation through gis techniques. International Journal of Climatology: A Journal of the Royal Meteorological Society, 20(14):1823–1841.
Rigobon, R. (2003). Identification through heteroskedasticity. Review of Economics and Statistics, 85(4):777–792.
Shen, H., Jiang, Y., Li, T., Cheng, Q., Zeng, C. and Zhang, L. (2020). Deep learning-based air temperature mapping by fusing remote sensing, station, simulation and socioeconomic data. Remote Sensing of Environment, 240:111692.
Singh, K. P., Gupta, S. andRai, P. (2013). Identifying pollution sources and predicting urban air quality using ensemble learning methods. Atmospheric Environ- ment, 80:426–437.
Wang, W. and Niu, Z. (2009). Var model of pm2.5, weather and traffic in los angeles–long beach area. In 2009 International Conference on Environmental Science and Information Application Technology, volume 3, pages 66–69. IEEE.