публикации ученых;regression analysis;hypothesis testing;greenhouse effect;global warming;climate change;atmospheric temperature
American Association for Science and Technolo-gy
Mukha, V. S. On the statistical methods for the conclusion on the climate change / V. S. Mukha // American Journal of Environmental Engineering and Science. – 2018. - Vol. 5. – No. 2. – Pр. 34 – 38.
Nowadays it is suggested, that the global warming is significant problem. There is a natural greenhouse effect which keeps the Earth warmer than it would otherwise be. This effect is enhanced by human activities and can lead to global warming and climate change. The Intergovernmental Panel on Climate Change (IPCC) publishes the reports of the state of knowledge on climate change in which there are also the instrumental evidences of the Earth warming. However, the scientific methods used for these conclusions are not mentioned. In this regard, there is a large number of “climate skeptics” who question both the fact of global warming and the role of the human in this process. In this work the importance of using instrumental scientific methods for the conclusions about climate change is emphasized. The analysis of the yearly mean value of the atmospheric temperature on the meteorological station Minsk (Belarus) over the last 20 years is considered. As a scientific method the statistical theory of regression analysis is used. At first the yearly mean values of the temperature on the base of the measurements of the temperature is calculated, then it is approximated by linear time-dependent regression function. Standard regression analysis procedures allow doing inference about significance of the linear dependence. In our case the linear stochastic dependence the yearly mean temperature on the time (empirical regression function) over the last 20 years has a slight tendency to increase. However, using the both statistical tests on the significant of the parameters and on the linearity of the regression function shows that the trend of the mean temperature is insignificant. The similar analysis shows that the linear approximation of the monthly mean temperature increases slightly for some months and decreases slightly for others, but these changes are insignificant.