Discharge data with high temporal resolution (1-10 days) is essential for many practical objections such as management of water resources, hydraulic structures engineering, and monitoring of coastal water quality. Although in a global scale there is lack of discharge data for most of rivers in the world. Nowadays, less than 60% of the world continental discharge is regularly measured at the river mouths. Regular direct measurements of discharge are performed only for a relatively small number of rivers, generally the biggest ones or ones that flow through densely populated areas. In some cases missing contact gauge measurements can be substituted by remote air and satellite measurements. In particular, earth-observing satellites provide effective instruments for global monitoring of river plumes which can serve as indicators of the river discharge. We have developed a new method of quantifying the volume of river discharge based on satellite observations of river plumes and numerical modeling. The general idea of the method is the following. Firstly, the spatial surface spread of the plume generated by the considered river discharge is identified using high resolution satellite imagery of the coastal zone adjacent to the river estuary. Secondly, a series of numerical simulations of the river runoff spread is performed under various prescribed external forcing conditions which include the discharge rate. Varying forcing conditions we iteratively improve the accordance between simulated and observed river plumes therefore consequentially specifying the value of river discharge. Practical importance of this work consists in the fact, that the suggested method is an alternative for the expensive and laborious direct measurements of the river discharge. The developed method was validated against in situ date for several rivers feeding the Black Sea. Then using this method we evaluated the variability of the river discharge inflowing from the Russian coast to the Black Sea for the period 2002-2010.