'Looping caves' versus 'water table caves': The role of base-level changes and recharge variations in cave development

- Looping caves are very common due to rapid valley incision and recharge fluctuations.
- Speleogenesis is modelled mathematically.
- Recharge fluctuations are shown to be much more important than previously thought.
- A speleogenetic model is proposed that replaces the one commonly in use.

The vertical organisation of karst conduit networks has been the focus of speleogenetic studies for more than a century. The four state model of Ford and Ewers (1978), which still is considered as the most general, relates the geometry of caves to the frequency of permeable fissures. The model suggests that the 'water table caves' are common in areas with high fissure frequency, which is often the case in natural settings. However, in Alpine karst systems, water table caves are more the exception than the rule. Alpine speleogenesis is influenced by high uplift, valley incision rates and irregular recharge. To study the potential role of these processes for speleogenesis in the dimensions of length and depth, we apply a simple mathematical model based on coupling of flow, dissolution and transport. We assume a master conduit draining the water to the spring at a base level. Incision of the valley triggers evolution of deeper flow pathways, which are initially in a proto-conduit state. The master conduit evolves into a canyon following the valley incision, while the deep pathways evolve towards maturity and tend to capture the water from the master conduits. Two outcomes are possible: a) deep pathways evolve fast enough to capture all the recharge, leaving the master conduit dry; or b) the canyon reaches the level of deep pathways before these evolve to maturity. We introduce the Loop-to-Canyon Ratio (LCR), which predicts which of the two outcomes is more likely to occur in certain settings. Our model is extended to account for transient flow conditions. In the case of an undulating master conduit, floodwater is stored in troughs after the flood retreat. This water seeps through sub-vertical fractures ('soutirages') connecting the master conduit with the deep pathways. Therefore, the loops evolve also during the dry season, and the LCR is considerably increased. Although the model is based on several approximations, it leads to some important conclusions for vertical organisation of karst conduit networks and stresses the importance of base-level changes and transient recharge conditions. It therefore gives an explanation of speleogenesis that relies much more on the dynamic nature of water flow than on the static fracture density.