Two seismic sources were tested on different surfaces during acquisition of a 5.5-km long high resolution seismic reflection (HRS) profile on the Ljubljana Moor in central Slovenia. Maximum target depth range of the survey was 200 - 300 m. Nine different combinations of source-surface conditions were analysed. Seismic sources included a seismic shotgun and an accelerated weight dropper system. The HRS profile crossed different surfaces including: paved roads, gravel roads, ploughed fields, grasslands and marshlands. Seismic source performance on different surfaces was evaluated through frequency and S/N ratio analyses of seismic data, and analysis of source-generated coherent noise appearance on seismograms. Results show that both seismic sources are suitable for some of the surfaces found in the surveyed area. The accelerated weight dropper produced the best results on gravel road surface, but poor results were found on grassland and ploughed fields. The seismic shotgun produced the best results in water-saturated soil on marshlands. It performed less favourably in unsaturated soil, generating more coherent noise. Water-saturated marshland surface and gravel road surfaces were found to be the most favourable for acquisition of high-resolution seismic reflection data on the Ljubljana Moor., Jure Atanackov and Andrej Gosar., and Obsahuje bibliografii