Abstract
As a first step toward understanding the marine litter balance in Tokyo Bay, we confirmed the applicability and accuracy of the two-way particle-tracking model（PTM）and the inverse method using a Lagrange multiplier developed by Isobe et al.（２００９） and Kako et al.（２０１０）for the East China Sea using high-frequency ocean surface radar-derived surface current velocities and a drifting litter collection database managed by the Ministry of Land, Infrastructure, Transport and Tourism. We focused on how the accuracy of these methods depends on horizontal diffusivity（HD）. We found that the two-way PTM and the inverse method are applicable to Tokyo Bay : the errors of identifying marine litter sources by the two-way PTM are within １０km, and the estimating marine litter inflows are０．９ to １．６ times the true inflows. The errors of identifying the sources depend on the HD :
the possibility of incorrect identification by the two-way PTM increases with larger HD. In addition, the errors depend on the strengths of the horizontal dispersion field produced by horizontal velocity shear. The possibility is also high for current fields with strong horizontal velocity shear. Also, when the HDs are larger than the corresponding true HDs, the inflows are overestimated. However, the dependence of the estimation on HD is not linear : if the HDs are larger than twice the true diffusivities, estimated inflows are less than twice the true inflows. Even if the HDs are one order of magnitude greater than the true diffusivities, the estimated inflows are smaller than six to seven times the true inflows.
