ASSESSMENT OF DROUGHT - USING NONLINEAR AGGREGATED DROUGHT INDEX IN ARTIFIICAL NEURAL NETWORK
| Author(s) | : | Ms. Chaitaly Joshi, Ms. Seema A Jagtap |
| Institution | : | ME (WRE- Civil engineering) Student YTCEM, Mumbai |
| Published In | : | Vol. 4, Issue 7 — July 2017 |
| Page No. | : | 554-560 |
| Domain | : | Engineering |
| Type | : | Research Paper |
| ISSN (Online) | : | 2348-4470 |
| ISSN (Print) | : | 2348-6406 |
Drought, is caused by the absence of rain is a naturalphenomenon. This has its hazardous impacts on humanlife, industries, environment, and community in general.Understanding drought is thus a very important processconsidering the present erratic rainfall scenario. It isobserved that Maharashtra state has been suffering fromlack of rainfall for over two decades, thus droughtassessment has become a serious concern in the area.Whenever a drought event and resultant disaster occur,governments and donors follow impact assessments, andresponse, recovery and reconstruction activities, toreturn the region or locality to a pre-disaster state.Therefore, it is well recognized that preparedness fordrought is the key to the effective mitigation of droughtimpacts which is becoming more important for waterresources managers to handle the challenges in waterresources management. To reduce such after eventrecovery, it is very essential that drought ischaracterized at the very initial stages and all therequired measures are taken beforehand to supplyproper water to all the people. There are many methodswith which one can assess drought. In this paper, we aimto develop drought forecasting tools by using software’ssuch as SPSS and ANN. The project aims to focus atManjhara river basin of Latur District in Maharashtra.A new drought index called Nonlinear AggregatedDrought Index (NADI) was established and a time seriesNADI time series was created using SPSS (StatisticalPackage for Social Science) software. In order todevelop NADI time series the raw data was convertedinto a series of transformed variables in order togenerate Principal components of various months. Thisdata was then put in ANN software and assessment wasdone.Using this time series, forecast up to 6 months oflead time is possible.
Ms. Chaitaly Joshi, Ms. Seema A Jagtap, “ASSESSMENT OF DROUGHT - USING NONLINEAR AGGREGATED DROUGHT INDEX IN ARTIFIICAL NEURAL NETWORK”, International Journal of Advance Engineering and Research Development (IJAERD), Vol. 4, Issue 7, pp. 554-560, July 2017.
