Application of Adjoint-based Forecast Sensitivities to Asian Dust Transport Events in Korea

Statistical adjoint sensitivity distributions of meteorological forecast errors of Asian dust transport events in Korea

“…Typically, a high-pressure system follows behind a surface frontal cyclone over the Korean Peninsula, accompanying the eastward movement of the 500 hPa westerly trough, during Asian dust events in Korea (e.g. Merrill and Kim, 2004;Kim et al, 2008b;Kim, 2008). Therefore, meteorological forecast errors affecting Asian dust transport forecasts in Korea are primarily associated with *Corresponding author.…”

“…Accurate predictions of meteorological fields are crucial for predicting Asian dust events because the air quality models for Asian dust prediction use predicted meteorological fields from numerical weather prediction models as input (Kim et al, 2008b). Such a requirement was demonstrated by Zhang et al (2007) when attempting to predict ozone concentrations in the vicinity of Houston, Texas, and by Liu et al (2011) when attempting to predict CO 2 transport around the globe.…”

“…However, the observational data over the source regions are insufficient to enable accurate Asian dust prediction. Moreover, the locations of the KMA observational sites over the dust source regions were determined rather subjectively (Kim et al, 2008b), specifically by installing stations following the known paths of Asian dust from the source regions to Korea. By investigating the sensitive regions of meteorological forecast errors to initial conditions for two Asian dust events in Korea, Kim et al (2008b) showed that not only dust source regions but also other upstream regions may be important to obtain correct atmospheric circulation patterns associated with Asian dust events and, thus, more accurate Asian dust forecasts.…”

“…Yumimoto et al (2008) applied an adjoint inversion of a four-dimensional variational data assimilation system of a regional dust model (RAMS/CFORS-4DVAR) to an Asian dust case by evaluating the impact of dust emission observations at specific sites. In contrast, by investigating adjoint sensitivities of atmospheric circulation forecast errors, Kim et al (2008b) demonstrated the feasibility of using adaptive observation strategies to design observation network for Asian dust events. Using adjoint-based forecast sensitivities to detect sensitive regions, Kim et al (2008b) showed that adaptive observations in sensitive upstream regions may be beneficial for short-range forecasts of two Asian dust events observed in Korea.…”

“…In contrast, by investigating adjoint sensitivities of atmospheric circulation forecast errors, Kim et al (2008b) demonstrated the feasibility of using adaptive observation strategies to design observation network for Asian dust events. Using adjoint-based forecast sensitivities to detect sensitive regions, Kim et al (2008b) showed that adaptive observations in sensitive upstream regions may be beneficial for short-range forecasts of two Asian dust events observed in Korea. However, to determine permanent or semipermanent observational sites to improve the forecasts and predictability of Asian dust transport events in Korea, adaptive observation strategies should be applied to numerous Asian dust cases to obtain statistically significant results.…”