A framework to identify homogeneous drought characterization regions
Drought monitoring is a complex phenomenon, as several climatic variables are required to accurately monitor and forecast drought. Furthermore, inappropriate existence of gauge stations scattered over the region without any comprehensive drought monitoring framework might end up to misleading conclusions. In this study, we aimed to develop a novel regionalized drought monitoring framework, which requires minimal drought monitoring stations. For this purpose, we considered K-means clustering algorithm based on the transient behavior to identify homogenous drought characterization regions. We applied our proposed framework on 52 meteorological stations across Pakistan in such way that each cluster consists of those meteorological stations that have a similar pattern with respect to the natural behavior of drought severity. We found nine meaningful clusters and the sum of square of the deviations in each cluster is very low. Further, the correlation within these clusters confirms the results of transition-based clustering method. The scatter plots and the Pearson correlation were used to assess the performance of the developed structure of homogenous drought characterization regions. Results show that instead of using individual observatory, any station located within the cluster can be considered for monitoring and forecasting drought of whole region. In summary, minimal and appropriate selection of optimal drought monitoring stations may incorporate to study overall regionalized behavior of drought. However, the choice of weather station depends on the existence of climatic parameters, its reliability, and historical availability of data on environmental variables.
The authors are grateful to the Pakistan Meteorological Department for providing data organized by the Karachi Data Processing Center.
The authors extend their appreciation to the Deanship of Scientific Research at King Saud University for funding this work through research group no. RG-1437-027.
- Ali Z, Hussain I, Faisal M, Nazir HM, Abd-el Moemen M, Hussain T, Shamsuddin S (2017a) A novel multi-scalar drought index for monitoring drought: the standardized precipitation temperature index. Water Resour Manag:1–13Google Scholar
- Atta-Ur Rahman (2016) Disaster risk reduction approaches in Pakistan. Springer Verlag, TokyoGoogle Scholar
- Brebbia CA (2011) The sustainable world (vol. 142). WIT PressGoogle Scholar
- Desikan P, Srivastava J (2004) Mining temporally evolving graphs. In: Proceedings of the sixth WEBKDD workshop in conjunction with the 10th ACM SIGKDD conference (vol 22)Google Scholar
- Dimtriadou E (2009) Cclust: convex clustering methods and clustering indexes. R package version 0.6–16, URL https://CRAN.R-project.org/package=cclust. Accessed Sept 2017
- Gui Y, Shao J (2017) Prediction of precipitation based on weighted Markov chain in Dangshan. In: Proceedings of the International Conference on High Performance Compilation, Computing and Communications. ACM, pp 81–85Google Scholar
- Hosking, J. R. M., & Hosking, M. J. (2017). Package ‘lmom’. Retrieved from https://www.r-project.org
- Huang JZ, Ng M, Ching WK, Ng J, Cheung D (2001) A cube model and cluster analysis for web access sessions. In: International workshop on mining web log data across all customers touch points. Springer, Berlin, pp 48–67Google Scholar
- McKee TB, Doesken NJ, Kleist J (1993). The relationship of drought frequency and duration to time scales. In: Proceedings of the 8th Conference on Applied Climatology, vol 17, no. 22. American Meteorological Society, Boston, pp 179–183Google Scholar
- Palmer WC (1965) Meteorological drought (Vol. 30). US Department of Commerce, Weather Bureau, Washington, DCGoogle Scholar
- R. Core Team (2014) R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, p 2014Google Scholar
- Scholz M (2016) R package clickstream: analyzing clickstream data with Markov chains. J Stat Softw 74(4)Google Scholar
- Shafer BA, Dezman LE (1982) Development of a surface water supply index (SWSI) to assess the severity of drought conditions in snowpack runoff areas. In: Proceedings of the western snow conference, vol 50. Colorado State University Fort Collins, pp 164–175Google Scholar
- Svoboda M, Fuchs B (2016) Handbook of drought indicators and indices. lincoln, national drought mitigation center. Google Scholar https://scholar.google.com/scholar_lookup?title=Handbook+of+Drought+Indicators+and+Indices&author=Svoboda,+M.&author=Fuchs,+B.&publication_year=2016
- Tigkas D, Vangelis H, Tsakiris G (2017) An enhanced effective reconnaissance drought index for the characterisation of agricultural drought. Environmental Processes 4(1):137–148Google Scholar
- Wilhite DA, Buchanan-Smith M (2005) Drought as hazard: understanding the natural and social context. In: Wilhite DA (ed) Drought and water crises: science, technology, and management issues. CRC Press, Boca Raton, pp 3–29Google Scholar
- Wing EAS (2010) Finance division. Government of Pakistan, “Pakistan Economic Survey”, Varios números, with Pakistan. (2007). Pakistan Economic survey. Islamabad: Economic Advisor's Wing, Ministry of Finance. Available: http://www.finance.gov.pk/survey_0708.html