| Next to the ocean, the terrestrial component of the climate system has a memory that controls the predictability of a climate at seasonal to longer time scales. However, the present understanding of the terrestrial memory is limited to the near-surface soil, and the role of deep-soil reservoir is unknown or has simply been neglected. Observational evidence shows the existence of plant roots much deeper in the soil/rock profile than the depth usually incorporated in existing land surface models. Moreover, plants in water-limited environments use a mechanism called "hydraulic redistribution" for transferring moisture via the root system from moist soil layers to dry soil layers. Analyses using observed and simulated datasets of soil moisture indicate an increasing memory with soil depth. Given that the deep-layer of the terrestrial component has a longer memory than the near-surface and that plants are the primary channels that link the soil environment with the overlying atmosphere, the roles of deep-rooting and hydraulic redistribution mechanisms may be far important than often thought. In this study, we developed a model that takes into consideration the deep-rooting and hydraulic redistribution and used it to assess the effects of these mechanisms on soil climatology and fluxes at the land-atmosphere interface. Our analyses suggest that hydraulic redistribution coupled with deep-rooting could be an important mechanism for establishing strong interaction between the long-memory deep-soil reservoir and the relatively short-memory atmospheric system. Thus, these mechanisms have important implications in long-term climate and ecological predictions, and may demand reevaluation of the modeling approaches used for root water uptake in vegetated environments experiencing water limitations. |
