Effects of drying on volcanic soil degradation: practical implication on hydrological behaviour

Many volcanic soils (Andisols) show an irreversible volume changing after drying. The irreversible changes occur when elementary unit particles, weakly bonded together, come sufficiently close to enable strong bonding; larger size aggregates are formed and they are not broken up on re-wetting. These modifications affect very much the type, volume and pore-sizes distribution and therefore modifies several physical properties (i.e. water retention, hydraulic conductivity and apparent hydrodynamic dispersivity) producing remarkable effects on many hydrological processes taking place in the soil. Several basic mechanism are not yet know on this distinctive effect: among others the energy level at which the irreversible change starts and develop, its development rate, and the effect of the time on the energy accumulation process. This deficit of knowledge limits our capability in distinguishing the status, in respect of the time scale, of the field degradation and therefore the prediction capability of the process. Such scenario is especially relevant considering that in much research and practical activities many physical properties are obtained on dried samples. One of the aim of this study is therefore the prediction of the hydrological behaviour of soil profiles studying the influence and the effects of samples drying. Two different case studies were investigated: the first one in an agricultural area with intensive tillage and inputs (SARNO plain); the second one on a natural slope (Monte FAITO). The drying effects on the Ap horizons of these different soil profiles have been compared in terms of the single physical property (i.e. water retention, saturated and unsaturated hydraulic conductivity, pore sizes distribution, PDF, etc..). Moreover, this drying effect on the solute and water balance has been tested. Undisturbed soil samples were collected in cylinders from the main horizons of volcanic soils classified as Humic Haplustand and Typic Hapludand. The Humic Haplustand was located in the SARNO plain, an alluvial plane surrounded by limestones covered by volcanic material; they were formed by pyroclastics fall and volcanic colluvial material. The Typic Hapludand soil was located at 1120 m a.s.l. on limestones relief covered by volcanic ash and pumices. Soil water retention and hydraulic conductivity functions were measured by means of tension table and Wind's method, saturated hydraulic conductivity by means of constant head permeameter and solute transport characteristics by means of a miscible flow experiment. These measurement were performed twice, the second one after the samples were oven dried. Main results of the comparison in the SARNO soil were that the drying induces: (i) reduction in the total porosity of the 20%; (ii) the translation of the soil water retention curve (at least for values of potential ranging between -5 and -350 cm), along with the corresponding pore-size distribution. Taking into account the invariance of the bulk density, this confirms that the porosity reduction occurs in the small pores region; (iii) an augmented fraction of large pores, as indicated by the pore-size distribution analysis; (iv) the saturated hydraulic conductivity values agree with the new distribution; (v) the breakthrough curve become slightly asymmetric and a preferential flow can be hypothesised. Main results of the FAITO soil were that the drying induces: (i) reduction in the total porosity of the 10%; (ii) this porosity reduction occurs in the small pores region. Furthermore, a physically based water balance model has been applied calculating a specific 'functional properties' for each case study. The effect of the drying on the Ap horizon was compared in terms of environmental risks. Particularly, in the SARNO plain the functional properties derived were (i) the solute resident time and (ii) an index of groundwater vulnerability while in the FAITO mountain was (iii) an index derived by the surface runoff.