Articles | Volume 8, issue 12
Atmos. Meas. Tech., 8, 5089–5097, 2015
https://doi.org/10.5194/amt-8-5089-2015
Atmos. Meas. Tech., 8, 5089–5097, 2015
https://doi.org/10.5194/amt-8-5089-2015

Research article 08 Dec 2015

Research article | 08 Dec 2015

Automatic volcanic ash detection from MODIS observations using a back-propagation neural network

T. M. Gray and R. Bennartz T. M. Gray and R. Bennartz
  • Department of Earth and Environmental Sciences, Vanderbilt University, Nashville, Tennessee, USA

Abstract. Due to the climate effects and aviation threats of volcanic eruptions, it is important to accurately locate ash in the atmosphere. This study aims to explore the accuracy and reliability of training a neural network to identify cases of ash using observations from the Moderate Resolution Imaging Spectroradiometer (MODIS). Satellite images were obtained for the following eruptions: Kasatochi, Aleutian Islands, 2008; Okmok, Aleutian Islands, 2008; Grímsvötn, northeastern Iceland, 2011; Chaitén, southern Chile, 2008; Puyehue-Cordón Caulle, central Chile, 2011; Sangeang Api, Indonesia, 2014; and Kelut, Indonesia, 2014. The Hybrid Single Particle Lagrangian Integrated Trajectory (HYSPLIT) model was used to obtain ash concentrations for the same archived eruptions. Two back-propagation neural networks were then trained using brightness temperature differences as inputs obtained via the following band combinations: 12–11, 11–8.6, 11–7.3, and 11 μm. Using the ash concentrations determined via HYSPLIT, flags were created to differentiate between ash (1) and no ash (0) and SO2-rich ash (1) and no SO2-rich ash (0) and used as output. When neural network output was compared to the test data set, 93 % of pixels containing ash were correctly identified and 7 % were missed. Nearly 100 % of pixels containing SO2-rich ash were correctly identified. The optimal thresholds, determined using Heidke skill scores, for ash retrieval and SO2-rich ash retrieval were 0.48 and 0.47, respectively. The networks show significantly less accuracy in the presence of high water vapor, liquid water, ice, or dust concentrations. Significant errors are also observed at the edge of the MODIS swath.

Download
Short summary
Volcanic ash poses a serious threat to aircraft traffic. A simple neural-network based technique was developed to detect volcanic ash from space using satellite infrared observations. A validation study shows promising results for several individual case studies. Issues remain near the edge of the satellite's field of view as well as in situations where ash is mixed with meteorological clouds.