Abstract
The present study examines the teleconnection of ENSO andIOD with monsoon rainfall (MN) and low, moderate and heavy rainfall events (LREs, MREs and HREs) over the Krishna river basin, using generalized additive models (GAMs) with suitable distribution. The outputs of GAMs indicate that, Poisson distribution is superior to the other distributions in assessing the teleconnection of ENSO-MN-IOD in the study area. Further, study resultsshowed that ENSO and IOD has significant (p < 0.001) non-linear responses to theLREs, MREs,HREs and MN. The influence of IOD on MN, LREs, MREs and HREs found positive on some parts, while negative on the other parts of the study area (i.e. heterogeneous in nature). While, ENSO has consistent negative influence on MN, LREs, MREs and HREsin the study area. Furthermore, La Niña and El Niño had positive and negative influenceon the MN, LREs, MREs and HREs respectively. The study outcomes will help the hydro-meteorologist and water related policy makers in modeling the impact of monsoon rainfall system on water, agriculture and allied sectors.
Highlights
The monsoon rainfall (MN) is the main source of water for the agriculture, domestic and industrial related activities in the Indian subcontinent
None of the past studies have recommended which distribution is best suitable tostudy the teleconnection of monsoon season rainfall with oceanic circulations El Niño Southern Oscillation (ENSO) and Indian Ocean dipole mode (IOD). This aspect has been examined in the current study, prior to the use of GeneralizedAdditive Models (GAMs) to assess the relationship between ENSO-MN-IOD, GAMs fitted with Gamma, inverse-Gaussian, Quassi, Quassi-Poisson, Gaussian and Poisson distributions and compared the model’s output
The comprehensive analysis of nonlinear influences of ENSO and IOD on rainfall events at different thresholds (LREs, Moderate rainfall events (MREs) and Heavyrainfall events (HREs)) and monsoon rainfall has been carried out using GAMs over Krishna river basin
Summary
The monsoon rainfall (MN) is the main source of water for the agriculture, domestic and industrial related activities in the Indian subcontinent. Krishnaswamy et al (2015) have used GAM to detect non-linear influence of IOD and ENSO on Indian summer monsoon rainfall (ISMR) / extreme rainfall events (EREs). None of the past studies have recommended which distribution is best suitable tostudy the teleconnection of monsoon season rainfall with oceanic circulations ENSO and IOD. This aspect has been examined in the current study, prior to the use of GAMs to assess the relationship between ENSO-MN-IOD, GAMs fitted with Gamma, inverse-Gaussian, Quassi, Quassi-Poisson, Gaussian and Poisson distributions and compared the model’s output. Based on the p-value and deviation explained, the best distribution was selected to detect non-linear association of IOD and ENSO with the monsoon rainfall and rainfall events at different thresholds in the study area
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
