In karst systems, near-surface dissolution of carbonate rock results in a high spatial and temporal variability of groundwater recharge. To adequately represent the dominating recharge processes in hydrological models is still a challenge, especially in data scarce regions. In this study, we developed a recharge model that is based on a conceptual model of the epikarst. It represents epikarst heterogeneity as a set of system property distributions to produce not only a single recharge time series, but a variety of time series representing the spatial recharge variability. We tested the new model with a unique set of spatially distributed flow and tracer observations in a karstic cave at Mt. Carmel, Israel. We transformed the spatial variability into statistical variables and apply an iterative calibration strategy in which more and more data was added to the calibration. Thereby, we could show that the model is only able to produce realistic results when the information about the spatial variability of the observations was included into the model calibration. We could also show that tracer information improves the model performance if data about the spatial variability is not included

The purpose of this research was to identify groundwater recharge mechanisms and measure fl ow velocities through the epikarst and the vadose zone in a typical site of Mediterranean carbonate karst in natural rain conditions. By avoiding artifi cial fl ushing more realistic travel times could be measured, and the effective resultant vadose fl ow velocity could be inferred. Additionally, detailed monitoring enabled better description and quantifi cation of infi ltration and percolation processes.Nine different cave drips were monitored and sampled for three hydrological years in fractured karst lithology on Mount Carmel, Israel. At three drips, discharge was measured continuously by tipping buckets. Rainfall and soil water content were recorded above the cave. An artifi cial tracer experiment was conducted using uranine placed in a joint and at the soil-rock interface at soil pockets, both 27 m above the cave.Four hydrological types of drips: post-storm, seasonal, perennial, and overfl ow, were identifi ed; each demonstrating different characteristics in terms of discharge and chemistry.The maximum effective fl ow velocities (uranine dye arrival times) were 41-76 cm/h; tracer mass fl ux was computed from the uranine breakthrough curves and drip hydrographs. The dominant fl ow velocities, derived from the peak of tracer mass fl ux, were 0·35-0·41 cm/h from post-storm drips, and 22-24 cm/h from the other drips.'Pulse-through' (intra-event piston fl ow) was indicated by the time lags between the increase in drip rate and the time of drop in salinity and appearance of the tracer in most of the measured drips. Chloride concentrations were used to distinguish between 'old/matrix water' and 'preferential-fl ow water' in the perennial drips, demonstrating a two-component mixing model. In 2005-2006 'preferential-fl ow water' reached up to 20-25% from annual discharge of these drips. This study promoted a method to compute dominant effective fl ow rates in the vadose zone in natural rain conditions.

Abstract:To investigate processes of water percolation, the drip response of stalactites in a karstic cave below a 143 m 2 sprinkling plot was measured. The experiment was conducted in Mount Carmel, Israel, at the end of the dry season and intended to simulate a series of two high-intensity storms on dry and wet soils. In addition to hydrometric measurements (soil moisture, surface runoff, stalactite dripping rates), two types of tracers (electrical conductivity and bromide) were used to study recharge processes, water origin and mixing inside a 28-m vadose zone. Results suggested that slow, continuous percolation through the rock matrix is of minor importance and that percolating water follows a complicated pattern including vertical and horizontal flow directions. While bromide tracing allowed identification of quick direct flow paths at all drips with maximum flow velocities of 4Ð3 m/h, mixing analysis suggested that major water fractions were mobilized by piston flow, pushing out water stored in the unsaturated zone above the cave. Under dry preconditions, 80 mm of artificial rainfall applied in less than 7 h was not enough to initiate significant downward water percolation. Most water was required to fill uppermost soil and rock storages. Under wet preconditions during the second day sprinkling, higher water contents in soils and karst cavities facilitated piston flow effects and a more intense response of the cave drips. Results indicate that in Mediterranean karst regions, filling of the unsaturated zone, including soil and rock storages, is an important precondition for the onset of significant water percolation and recharge. This results in a higher seasonal threshold for water percolation than for the generation of surface runoff.

Arbel, Y., Greenbaum, N., Lange, J., Shtober-Zisu, N., Grodek, T., Wittenberg, L., and Inbar, M. 2008. Hydrologic classification of cave drips in a Mediterranean climate, based on hydrograph separation and flow mechanisms. Isr. J. Earth Sci. 57: 291-310.Processes of infiltration to groundwater in a karstic area were studied by monitoring and sampling cave drips during 2004-2008 at two sites with different lithologies: dolomite of the Yagur Fm. and crystalline limestone of the Muhraqa Fm., in Mt. Carmel, Israel. Two tracer tests under different antecedent moisture conditions and "rainfall" intensities were performed.At both sites, 4 hydrological drip types were identified: Post-storm, Seasonal, Perennial, and Overflow, each exhibiting different patterns of discharge, chemistry, and travel time. Perennial drips represent the slow ("matrix") component; however, discharge and [Cl -] fluctuations after intense rainstorms indicate the relative contribution of by-pass flows (event-water). Based on [Cl -] during the natural season and electrical conductivity (EC) during the sprinkling experiment, seasonal and storm input of event-water versus old-water were calculated in the various drips using hydrograph separation. The fractions of calculated event-water in the perennial drips were always <30%. Overflow drips started late in the season, after drainage in the nearby drips exceeded their discharge capacity.Hydrograph recessions of drips have exponential drainage of few "reservoirs". Perennial and seasonal drips had at least two recession segments: (1) recession of quick-flow governed by a piston flow effect and small input of preferential flow, which lasted up to 10 days; (2) slow drainage of the vadose zone, which continued for a few months or until the next season. The recession constant(s) for post-storm and overflow drips were shorter than 10 days."Piston flow" effect indicators were the lag time of injected tracers after hydrographs onset, and the larger component of old-water in drips-hydrograph onset and rising limb. The different processes, associated lag times, and flow velocities have major impacts on groundwater vulnerability and aquifer recharge.

Dryland areas are regarded as highly sensitive to climatic changes. A positive relationship between rainfall and environmental factors is often assumed for areas with an average annual rainfall of 100–300 mm. This assumption disregards the fact that a climate change in arid areas is not limited to climatic factors. It is often accompanied by a pronounced spatial variability in surface characteristics. The present work deals with the complex relationships among average annual rainfall, surface properties and the spatial redistribution of water resources in sandy areas located in the Northern Negev Desert. Two case studies are considered. The first deals with the hydrological effects of biological topsoil crusts on the water regime, along a rainfall gradient (86–170 mm). This study is based on five monitoring sites. Data obtained show a decrease in water availability with increasing annual rainfall. The findings are attributed to the decisive role played by the non-uniform properties of the topsoil crust along the rainfall gradient. The second case refers to the non-uniform development, and survival, of planted trees. Trees planted on steep dunes are well developed, with a high survival rate, whereas trees planted on low angle dunes are small. This study focused on the role of a water repellent layer on the water regime. Data obtained show a striking difference between steep and low dunes in all aspects studied, namely the degree of water repellency, frequency and magnitude of runoff events, infiltration depth and soil moisture. All variables monitored were found higher on steep than on low dunes. The large trees shed a substantial amount of leaves, whose decay developed a water repellent layer. Runoff generation over the repellent layer enhanced deep water penetration, through the process of subsurface flow. The lack of a water repellent layer over the low dunes prevented runoff generation, with its positive effects.