[S-VC29] Volcanic and igneous activities, and these long-term forecasting

[SVC29-P11] Analysis of distance decay patterns of tephra fall layer thicknesses: Characteristics of point clouds with relationships between straight-line distances from a vent and tephra fall layer thicknesses

*Shimpei Uesawa¹, Shingo Takeuchi¹, Kiyoshi Toshida¹ (1.Central Research Institute of Electric Power Industry)

Keywords:Tephra fall thickness, Isopach map, Digital tephra fall distribution map, Exponential decay function, Power-law

Tephra fall thickness maps are important for (1) estimating the amount of tephra fall and (2) understanding the dynamics of explosive volcanic eruptions based on the characteristics of tephra fall distribution. Traditionally, isopach or isopleth maps have been created by plotting layer thickness or mass per unit area point data collected by geologists on a map by hand. In recent years, geologists have attempted to create distribution maps using computer calculations (e.g., González-Mellado and De la Cruz-Reyna, 2010; Green et al., 2016; Yang and Bursik, 2016; Tajima, 2021). These methods may make us easy to obtain a distribution map immediately after an eruption, and we could use it to predict eruption trends. Additionally, such digital tephra fall distribution maps can be applied to the evaluation of tephra fall hazards by history-based hazard curves created with a tephra fall distribution database (Uesawa et al., in review). To develop the digital tephra fall database, which is a re-digitization of Sudo et al. (2007), we attempt to improve the computational model of creating a tephra fall distribution map. The short research history of the isopach map generation model with observation point data for tephra fall thickness is as follows. González-Mellado and De la Cruz-Reyna (2010) studied the decay of tephra layer thickness with distance from the vent at different angles from the wind direction for 14 eruptions and the semi-empirical tephra fall distribution model that incorporates the effect of diffusion was proposed. Subsequently, Yang and Bursik (2016) proposed an empirical interpolation model for tephra fall distribution assuming that the straight-line distance from the vent and the distance from the crater along the major axis of the tephra fall distribution (downwind distance) with the tephra fall layer thickness follow exponential decay function (trend model) as first-order approximations. Furthermore, the residuals with prediction values of the trend model are analyzed by a numerical prediction model based on geostatistical methods (kriging) and the final predicted values are corrected by adding the predicted residuals to the trend model to reproduce the complex distribution of tephra fall. In addition, Yang et al. (2019) used both exponential decay function and power-law to infer the location of the vent from the tephra layer thickness and tephra particle size in an inverse analysis, and the power-law model is based on González-Mellado and De la Cruz-Reyna's (2010) simplified term for the diffusion of tephra fall. Thus, it is clear that the distance decay property of the tephra fall layer thickness is the most basic information for the interpolation modeling for the isopach maps. To understand the diversity and general characteristics of the distance attenuation of the tephra layer thickness, we digitized the tephra fall point data with the published tephra fall distribution maps for six different scale recent cases in Japan and abroad using GIS and investigated the relationship between the tephra fall layer thickness and the straight-line distance from the vent as a point cloud. As a result, the attenuation characteristics of the point cloud upper limit were classified according to the scale, style, and distance from the vent as follows: a) Plinian to sub-Plinian eruptions show exponential or power-law attenuation in the proximal area and power-law attenuation in the distant region from the vent. b) Vulcanian (1 magmatic, and 2 phreatic) eruptions show power-law attenuation. In the case where we could track the tephra fall to a distant region, the upper limit of the distribution of the point cloud is approximated by two straight lines in the double logarithmic graph of the layer thickness versus the distance from the crater, regardless of the scale. These results suggest that it is necessary to select the appropriate function for the tephra layer thickness-distance attenuation characteristics when modeling them because the attenuation characteristics depend on the scales of the eruption and the distance from the vent. We will analyze more eruption cases to find a general relationship between eruption size, eruption style and, characteristics of tephra layer thickness attenuation for further considerations.