Volcanic radiative forcing reconstruction is an important part of paleoclimate simulation and attribution efforts, and the conversion factor used to transfer ice core-based sulfate observation into stratospheric volcanic aerosol loading (LTD factor) is critical for such reconstruction. A Pinatubo-based LTD was proposed and adapted in the CMIP5 and CMIP6 volcanic forcing, under the assumption that all tropical eruptions follow the same atmospheric transport and deposition pattern. This study revisits the LTD factor using a large collection of polar ice core records of Tambora deposition and a Monte Carlo sampling model. A new set of LTDs with associated uncertainties are obtained, which is in approximate with our previous Pinatubo-based LTD estimation in Greenland, while corrects the bias of over-representing the west Antarctic. The uncertainties revealed from the Monte Carlo simulation suggest that, difference in the ice core abundance only introduce limited uncertainty in LTD for individual eruption, once reach a certain threshold (about 15 in Greenland and 20 in Antarctic). The comparison of Southern Hemispheric LTD among Tambora, Agung, and Pinatubo suggests that, the conversion factor may vary for individual eruption. Results obtained from this study may ease our conventional proneness to use as much ice core observations as available to estimate icecap volcanic deposition, while emphasize the importance to build a distribution of the LTD ideally for eruptions with different size and locations. Meanwhile, the estimated sets of Tambora-based LTDs could serve as a compromising choice in future volcanic forcing reconstruction work, especially when Tambora is utilized as a reference

Proxy-model comparisons show large discrepancies on volcanic aerosols’ hydrological effects in the Asian monsoon region (AMR). This was mostly imputed to uncertainties of the single model used in previous studies. Here, we compared two groups of CMIP5 multi-model ensemble mean (MMEM) with the tree-ring-based reconstruction Monsoon Asia Drought Atlas (MADA PDSI), to examine their reliability on reflecting hydrological effects of the volcanic eruptions in 1300-1850 CE. Time series plots indicate that MADA PDSI and MMEMs agree on the significant drying effects of volcanic perturbation over the monsoon-dominated subregion, while mismatches exist over the westerlies-dominated subregion. Comparisons on spatial patterns suggest that MADA PDSI and MMEMs agree better in one year after the volcanic eruption than in the eruption year, and in subregions with more available tree ring chronologies. MADA PDSI and CMIP5 MMEMs agree on the drying effect of volcanic eruptions in western-East Asia, South Asian summer monsoon and northern East Asian summer monsoon (EASM). Model results suggest significant wetting effect in southern EASM and western-South Asia, which agrees with the observed hydrological responses to 1991 Mount Pinatubo eruption. Analysis on LME model simulations show similar hydrological responses. These results suggest that CMIP5 MMEM is able to reproduce volcanic eruptions’ hydrological effects in southern AMR.