Research Paper | Statistics | Tanzania | Volume 6 Issue 11, November 2017
The Changing Nature of Probability Distribution of Daily Temperature: A Case of Coastal Zone of Tanzania
Abstract: 54 year datasets (1961 -2014) of recordings of the maximum and minimum daily (24 h) temperature in the coastal zone of Tanzania are analysed to test whether there is any significant (in statistical sense) trends in variability of daily temperature. In recent studies it has been reported that there are trends in mean temperature and other climate indices over Tanzania and many other countries worldwide. However, in Tanzania it is not clear whether there is also a trend in variability of daily temperature. In this study 3 meteorological stations (Dar es Salaam, Mtwara, Tanga) have been analysed for trends in variability using F-ratio test, least square method and Mann-Kendall trend test. The results indicate that the observed increasing/decreasing trends in variability of the maximum temperature are statistically non-significant at the level of 5 % while the trends are statistically significant for minimum temperatures. Daily minimum temperatures for Dar es Salaam and Mtwara show an increasing trend in variability while a decreasing trend for Tanga station has been observed.
Keywords: Variability, F-ratio test, Mann-Kendall trend test, Temperature, Tanzania
Edition: Volume 6 Issue 11, November 2017,
Pages: 1511 - 1517
How to Cite this Article?
Edwin C. Rutalebwa, "The Changing Nature of Probability Distribution of Daily Temperature: A Case of Coastal Zone of Tanzania", International Journal of Science and Research (IJSR), Volume 6 Issue 11, November 2017, pp. 1511-1517, https://www.ijsr.net/get_abstract.php?paper_id=ART20178261
How to Share this Article?
Similar Articles with Keyword 'Variability'
Meta-Analysis on the Effect of Tamiflu in Patients Infected with Flu in Some Areas around US and Asia
Identification of the Productive Potential of Customers in a Commercial Bank Portfolio: Calculation of the Customer Lifetime Value using Statistical Learning Method