We present high resolution paleosecular variation (PSV) records up to 8 cal. kyr BP from three piston cores, MD161/8, MD161/11 and MD161/13 acquired in the Krishna-Godavari (KG) basin, Bay of Bengal. During the Holocene period, high sedimentation rates are recorded at MD161/8 (38.8-248.3 cm/kyr), MD161/11 (137-336 cm/kyr) and MD161/13 (similar to 573 cm/kyr). Rock magnetic data analysis suggests that the remanence signal is carried by titanomagnetite/titanohematite grains in stable single domain (SSD)/pseudo single domain (PSD) state. The PSV records of MD161/11 and MD161/13 show good correlation in the uppermost sediments despite significant variation in the sedimentation rates; however, poor correlation of PSV records is observed for the core MD161/8 probably due to local effects. Paleoinclination records of MD161/8, MD161/11 and MD161/13 show a low between similar to 2.4 and 2.0 cal. kyr BP, an increase between 2.0 and 1.4 cal. kyr BP and a decrease towards the present. To varying degrees these trends can be observed in the other Asian PSV records of Shuangchiling (SCL) and Biwa lakes. However, the magnitude of the observed inclination anomaly in KG basin is higher (similar to 40 degrees) compared to those reported from SCL (similar to 25 degrees) and Biwa (similar to 10 degrees) lakes. Paleodeclination records of MD161/11 and MD161/13 show a decline between similar to 4.0 and 2.9 cal. kyr BP, an increase between 2.9 and 2.1 cal. kyr BP, a substantial decrease between similar to 2.1 and 1.0 cal. kyr BP and an increase towards the present. Similar trends can be observed in the other Asian PSV records of SCL and Biwa lakes with a minor age offset of 0.2-0.5 kyr. The available models CALS7k.2 and CALS10k.1 are evaluated for their capability in predicting the inclination and declination anomalies from the Asian regions. The CALS7k.2 model can predict most of the inclination anomalies while the CALS10k.1 is unable to predict many of them. The CALS7k.2 model shows that the observed inclination anomalies can be attributed to the presence of a strong non-axial dipolar magnetic field over the Bay of Bengal. More PSV records need to be constructed to better understand the anomalous non-axial dipolar field over the Bay of Bengal. (C) 2016 Elsevier B.V. All rights reserved.