Intensity of the Geomagnetic Field over the Past 3000 Years from Indian Subcontinent: Comparison with Global Geomagnetic Models and Other Data from Selected Regions

India is the only country that has data on the past geomagnetic field (GMF) strength in the entire South Asia, therefore construction of new regional intensity secular variation curve is essential to describe the past Earth magnetic field behavior in this region. Two palaeo secular variation (PSV) curves for the last 3000 years are proposed; one for the whole Indian Subcontinent and the other for the Southern India using data respectively within 1600 km and 1000 km radius circles. The Subcontinent and the Southern India data are relocated to their geographically centered reference points (Karaundi village and Bangalore city respectively). All data were weighted based on the thermal intensity protocol, the number of samples per site and the type of studied materials. Selected data with weight (>= 6 and >= 8) were used to calculate the intensity PSV curves using the sliding window technique. The converted Virtual Axial Dipole Moment (VADM) PSV curves of both data sets are similar with respect to age, but values of the Southern India are little higher than those for the Indian Subcontinent with differences being somewhat conspicuous similar to 400 BCE and between 1100 and 1600 CE. Such difference can be due to (i) input data difference for the two curves and/or (ii) varying geomagnetic effects between the Indian Subcontinent and South India. The intensity PSV curve also indicates that the maximum PSV rate of change per century is observed between 900 CE and 1000 CE time window (similar to 8 mu T/century for 1600 km; 10 mu T/century for 1000 km radius circles). Intensity PSV predictions by global geomagnetic models (ARCH10k.1, CALS10k.2, SHAWQ2k and SHAWQ-Iron Age and ArchKalmag14k.r) for India do not reflect the regional intensity PSV pattern obtained in the current study and thus suggest that regional GMF curves are more meaningful rather than global models to describe regional/local variations. The converted VADM PSV curve trend of the Indian Subcontinent, considering the error in VADM and age, closely agree in short time spans with Central Asia (Uzbekistan), Western and Eastern Europe, and Mexico VADM PSV curve trends but shows no match at all with the VADM patterns of Eastern Asia (China, Japan and Korea), Central-Western Europe. The differences in regional VADMs behavior could be mainly due to variable non-dipolar contributions. The present results provide a good opportunity to describe some of the main features of the upper Holocene (0-3000 Yrs) GMF for the Indian Subcontinent and it needs further refinement as the present analysis is based on limited data.