Usually recharge is calculated as the unknown parameter in a catchment water balance, the known parameters being rainfall (measured), ET (estimates by a variety of ways), surface inflow and outflow (measured if applicable), and the change in storage (from lake level and bathymetry). Ideally one would have daily data on all these parameters for a year, and that way calculate monthly water balances to get an idea of the seasonal variation in net recharge.

Thanks Saha, but the challenge before me is that, none of the climate/hydrological data is available except the rainfall. The soil recharge rate, can assume the generalized value for that soil type, the catchment area and rough land use pattern also can estimate. No other field data available.

Yes thats the challenge more often than not. One can cut down on the data required to just rainfall, ET and water level, and then setting the difference to include net inputs (surface runoff + recharge).

For evapotranspiration, one can approximate for a lake by open pan evaporation data, but if thats not possible to obtain, maybe one can either use 1 km MODIS ET data, or look up IndiaWaterPortal, they have a link to some met data files for each district of India. One can use Hamon eqn to estimnate ET from air temp, or Penman Monteith if one has net radiation, windspeed and RH.

But without water level data, one has no way of knowing how much is the net inpout or output into the lake, and any computation then becomes meaningless.

I've heard there are remote sensing methods that can deduce change in lake volume from changes in lake surface area.

how about considering the runoff volume (SCS CN) as the input volume ? As, the requirement is rough approximation of the capacity, I adopted it. I know it wont be exact, but won't it give a idea? Anyway, that issue is over, but still I like to find the solution

Please find an extract from my PhD thesis uploaded in RG

"4.3.6.4 Recharge from irrigation scheme

All the 6 medium and 40 minor irrigation schemes within the study area have a maximum head of water less than 3m. Hence the percolation of water is calculated either as 0.005m/day/plan area (Rushton, 1990) or 0.5% of the volume stored monthly (Ponrajah, 1984). Hence a range of 5% to 15% of volume stored was taken as practical range for the optimization models.

The capacity and irrigable area of all the irrigation schemes together with their percentage contribution within each polygon is shown in Table 4.12. Paddy and OFC cultivation in medium irrigation schemes and ASC division are given in Table 4.13"

This will give you a good guide to you

. I have uploaded three publications also in RG you can refer them also