Reply to comment by Y. Yamamoto on “Experimental reassessment of the Shaw paleointensity method using laboratory‐induced thermal remanent magnetization”

[1] Pan et al.'s [2002] aim was to investigate the reliability of the Shaw paleointensity method [Shaw, 1974; Kono, 1978; Rolph and Shaw, 1985] using basalt samples containing laboratory-induced thermal remanent magnetization as the natural remanent magnetization (NRM). The Shaw method uses alternating field (AF) demagnetization of the NRM and an induced thermal remanent magnetization (TRM) to allow determination of the NRM/TRM ratio as a function of coercive force. This method requires a single heating to above the Curie temperature to gain a full TRM. Further as a check for alteration during heating anhysteretic remanent magnetization (ARM) given before and after heating are compared. The main conclusions from the Pan et al. [2002] study are as follows: (1) The Shaw method yields correct paleointensity estimates for samples that show limited alteration, (2) rock-magnetic property investigations are important for sample selection to determine the highest laboratory temperature needed to impart the TRM and also to evaluate the reliability of estimations, and (3) the individual correction by ratios of anhysteretic remanent magnetization before and after heating (ARM1/ARM2) is a powerful approach to correct the TRM capacity change caused by the laboratory heating. Yamamoto [2003] has written a very useful commentary that concerns conclusion 3 and the nature of the TRM correction used. He also comments on the definition of the alteration index P. The comments are discussed below and the data recalculated, however the conclusions of Pan et al. [2002] remain unchanged. [4] Yamamoto also commented on the definition of P value. To quantify linearity discrepancies between TRM-ARM2 and NRM-ARM1, Pan et al. [2002] previously defined a P value as the percentage difference of residual magnetization of ARM2 and ARM1 after AF demagnetization at 150 mT. It is acknowledged that the residual ARM1 (ARM2) remaining after maximum demagnetization is inherited from incomplete demagnetization of NRM (TRM) as the maximum AF demagnetization field is the same peak field value used for ARM production. The P value is redefined as (Tr/TRM-Nr/NRM) × 100%. The new P value is shown in Table 1 and for this sample set does not produce significantly different values. [5] The main conclusions of Pan et al. [2002] remain unchanged. The ARM ratio correction is a powerful tool when alteration is limited and the P value is small; however, caution should be taken when the P is larger than 10%. We also stress the importance of investigating rock magnetic properties to select thermally stable samples and to determine the appropriate maximum temperature (Tc + 20°C) to produce the TRM.