Gully morphology (often summarized as comprising an alcove, channel and debris apron) is one of the key elements used to support the argument for liquid water in the recent past on Mars. Nevertheless, the processes that create different gully morphologies, both on Mars and on Earth, are not fully understood. One of the puzzling morphologic attributes of martian dune gullies is their apparent lack of an apron, or terminal deposit, which has caused debate about their formation process. Several physical processes could explain the formation of these gullies (e.g. runoff, debris flows, granular flows, sliding blocks falling downslope). In this work, we focus on the role of liquid in the substrate as well as in the flow and choose to experimentally test the plausibility of this hypothesis. We performed a series of analog experiments to investigate the formation of gullies on sand dune-like substrates. We used controlled flows of water over an inclined sand-box to produce gully-like forms. Ice-rich sedimentary substrates were used, including substrates that included a thin liquid water-saturated thawed layer (an "active layer") above the ice-saturated zone to give an analogue for a "periglacial" environment. We quantitatively demonstrate that debris flow processes in "periglacial" experiments are conducive to the formation of narrow and long channels with small terminal deposits with perched channels. By re-analysis of martian elevation data for dune-gullies on Mars, we have found good evidence that such terminal deposits could exist. Our experiments revealed that increased water content in the thawed layer above the frozen bed increases flow-length due to the subsequent reduction in