A new archive of large volcanic events over the past millennium derived from reconstructed summer temperatures

Information about past volcanic impact on climate is mostly derived from historic documentary data and sulfate depositions in polar ice sheets. Although these archives have provided important insights into the Earth’s volcanic eruption history, the climate forcing and exact dating of many events is still vague. Here we apply a new method of break detection to the first millenniumlength maximum latewood density reconstruction of Northern Hemisphere summer temperatures to develop an alternative record of large volcanic eruptions. The analysis returns fourteen outstanding cooling events, all of which agree well with recently developed volcanic forcing records from high-resolution bipolar ice cores. In some cases, however, the climatic impact detected with our new method peaks in neighboring years, likely due to either dating errors in the polar ice cores or uncertainty in the interpretation of atmospheric aerosol transport to polar ice core locations. The most apparent mismatches between forcing and cooling estimates occur in the 1450s and 1690s. Application of the algorithm to two additional and recently developed reconstructions that blend maximum latewood density and ring width data reproduces twelve of the detected events among which eight are retrieved in all three of the dendroclimatic reconstructions. Collectively, the new estimates of volcanic activity with precise age control provide independent evidence for forcing records during the last millennium. Evaluating the cooling magnitude in response to detected events yields an upper benchmark for the volcanic impact on climate. The average response to the ten major events in the density derived reconstruction is −0.60 °C± 0.13 °C. Other last millennium temperature records from proxies and model simulations reveal higher cooling estimates, which is, to some degree, related to the very different high frequency variance in these timeseries. 7 The Global Volcanism Program lists 259 total events between 1000 CE and the present with VEI greater than or equal to 4 for the Northern Hemisphere and the tropics. Among these, 204 are dated using historical observations. Other dates are mainly derived from radiocarbon and tephrochronology. Introduction Knowledge about past volcanism improves our understanding of the sensitivity of the climate system to exogenous radiative forcings. Associating large volcanic eruptions with climate anomalies and estimating their magnitude of impact requires a comprehensive and well-dated record of past volcanism (Esper et al 2013a). The majority of known large eruptions in the past millennium have been dated by © 2017 IOP Publishing Ltd historical observations (Global Volcanism Program 2013). In a climatological context, however, the amount and character of volcanic aerosols dispersed into the atmosphere are of greatest consequence and not well characterized by historical observations. In Environ. Res. Lett. 12 (2017) 094005 particular, radiation absorbing sulfate aerosols injected into the stratosphere undergo multiyear transport and distribution resulting in significant alterations of the earth’s energy budget (Robock 2000). In contrast to historical observations, volcanic particles deposited in polar ice sheets have been successfully used to estimate the amount and composition of sulfate aerosols from volcanic eruptions before the onset of modern observations. Combined with a model for atmospheric dispersion, ice core deposition records are used to derive radiative forcing reconstructions to evaluate the impacts of volcanic eruptions in climate model simulations (Schmidt et al 2011). A concern for ice core derived estimates of volcanic events is their potential for dating errors due to uncertainties in the age-depth models or false assignment of reference horizons to certain eruptions (Baillie and McAneney 2015, Sigl et al 2015). Even for a correctly dated ash layer, the time lag between ash injection, atmospheric perturbations and polar deposition can be up to 2.5 years and therefore can result in misinterpretations of environmental effects (Plummer et al 2012). Irregular snow accumulation causes high spatial variability in the amount of deposited volcanic material and, hence, additional large uncertainties in the magnitude of sulfate records (Hegerl et al 2006, Sigl et al 2014). Finally, translating the sulfurous ash deposits into radiative forcing estimates requires several physical assumptions about the character of aerosols, their visual properties and atmospheric transport, and thus again adds further uncertainty to the quantification of volcanic history (Toohey et al 2016). These shortcomings motivate the need for alternative and independent approaches to reconstructing the timing and climatic impact of volcanic eruptions. Previous studies of annually resolved climate proxies (Briffa et al 1998a), model simulations (Atwood et al 2016) and even early observational data (Jones et al 2003) have revealed volcanic eruptions to cause severe surface temperature cooling at hemispheric scales. The pulse-like forcing of these events triggers structural breaks (shifts in the mean) in time series of hemispheric or global mean temperatures, the magnitude of which exceed the natural range of climate variability (Naveau et al 2003). Our study is therefore motivated by the assumption that detecting and separating such breaks without prior knowledge of their occurrence will generate independent evidence of past volcanism constraining the timing and climatic effects of large eruptions. We use a new reconstruction of mean Northern Hemispheric extra tropical summer temperatures derived from maximum latewood density (MXD) records (Schneider et al 2015) to test this hypothesis. Some recent work has suggested that tree-ring records smooth and underestimate the cooling associated with volcanic eruptions (Mann et al 2012). But these findings were rebutted by numerous studies that 2 demonstrated a distinct volcanic signal in MXD data (Anchukaitis et al 2012, D’Arrigo et al 2013, Esper et al 2013a). In contrast, temperature reconstructions based on tree-ring width, the far more abundant dendrochronological parameter, were shown to underestimate abrupt cooling events due to biological memory effects (Esper et al 2015). The Schneider et al (2015) reconstruction is the first purely MXD-based record covering temperatures over more than a thousand summers and thus represents an ideal record for large eruptions. In this paper, we detect volcanic-induced breaks in this reconstruction independent of ice core estimates using an indicator saturation method. This technique has been described and validated in a pseudoproxy context (Pretis et al 2016) and is used to construct an independent chronology of volcanic eruptions that is then compared with existing ice core derived forcing records. We conclude by discussing the implications of our alternative record of past volcanism and evaluate our findings in the context of additional proxy reconstructions and climate model simulations.