The surface horizontal diffusivity are quantified using submesoscale [{\it O}(1) km spatial scale and {\it O}(1) hour time scale] observations of high-frequency derived surface current fields and geostationary ocean color imagery-derived surface chlorophyll concentrations in a coastal region. The given observations of the surface currents and chlorophyll concentrations are feeded into the advection-diffusion equation as the advection and concentration terms. We practice the ways to estimate the diffusion coefficients using an idealized flow model and the model-derived concentrations. The model currents are simulated uni-directional steady currents under specific temporal and spatial decorrelation scales with a Gaussian noise, and the concentrations are derived using random walk and flight schemes, which have been used for tracking of the concentration of phytoplankton, zooplankton, and individuals of fishes as a statistical approach. Then, the diffusion equations are solved in terms of random parameters and diffusion coefficients. This work will be applicable to the ecosystem process studies at submesoscale and improve the limitation of the present-day statistical modeling.