Efficient subsurface drainage design requires reliable estimates of hydrological loading reaching agricultural soils. This study investigates how two alternative representations of hydrological recharge, precipitation and percolation, affect subsurface drain spacing at parcel level. An integrated modelling framework was developed and applied to the Xirias River Basin in Thessaly, Greece, an intensively cultivated area characterized by heterogeneous soil and hydrological conditions. The Soil and Water Assessment Tool (SWAT) was used to simulate hydrological processes at the Hydrological Response Unit (HRU) scale. A GIS-based spatial linkage was then developed to transfer HRU-level hydrological information to individual agricultural parcels and integrate it with parcel specific soil hydraulic properties and drainage design parameters. Optimal subsurface drain spacing was calculated using the steady-state Donnan–Hooghoudt equation under two scenarios: precipitation-based recharge and SWAT-simulated percolation below the root zone. The comparison revealed substantial differences in spacing between the two approaches. Mean drain spacing increased from 38 m under the precipitation based scenario to 52 m when percolation was used, corresponding to a 36.8% increase. Spatial variability was further controlled by soil hydraulic properties, highlighting the importance of parcel specific conditions in drainage design. The proposed framework demonstrates that the representation of hydrological recharge can substantially influence calculated drain spacing. By linking watershed level hydrological simulations with parcel level drainage calculations, the approach provides a spatially explicit basis for more targeted and cost-effective agricultural drainage planning at large scales.