Characterization of co2 exchanges in deep soils and caves and their role in the net ecosystem carbon balance

Resumen

This Thesis arises to response to the anomalous CO2 fluxes detected by the Eddy towers located on two carbonate and semiarid sites in the south-east of Spain. In the soil-atmosphere CO2 exchanges in both El Llano de los Juanes in the Sierra de Gádor and also Balsa Blanca located in the Natural Park of Cabo de Gata-Nijar (Almeria province), significant CO2 emissions to the atmosphere were detected. These emissions could only come from the soil and not from the vegetation, occurring as they did mainly during the summer, with senescent vegetation. For this reason, this Thesis focuses on monitoring and characterization of the soil, distinguishing which are the main factors implicated in CO2 exchanges with the atmosphere. For the characterization and monitoring of subterranean CO2 soil CO2 profiles were established in both experimental sites with sensors installed at three different depths (0.15, 0.5 and 1.5 m). Furthermore, in El Llano de los Juanes sensors were installed at 0.25 m depth and in a borehole of 7 m depth, which was sealed from the surface to simulate a karst cave. With the first CO2 data obtained from the borehole, we realized that the deep soil could store large amounts of CO2, because in the first 7 meters we recorded values of up to 18000 ppm. Since we were interested in determining the factors involved in CO2 ventilation from soils and particularly in karstic areas containing numerous cavities, we launched an examination of the literature regarding cave ventilation. This review triggered the first publication of this Thesis (Chapter 1), because we realized that to determine the buoyancy of air masses inside and outside the cave, and therefore its ventilation, scientists were treating both air masses as if they had equal composition. However, both our measurements in the borehole and many others scientists working in other places had found great concentrations of CO2 in soil air, demonstrating a clear difference between the compositions of soil and atmospheric air. These realizations led us to develop formulas to correctly determine the virtual temperature and hence buoyancy of the air masses considering the weight of CO2 in the composition of both air masses. In Chapter 2 of this thesis, we applied the formulas previously developed for 14 caves and holes around the world. We demonstrate the significant errors that occur when predicting cave ventilation if not taking into account the effect of CO2 air composition. To clarify the calculation tasks, we provided an on-line tool to calculate the virtual temperature from input data comprised of CO2, air temperature and relative humidity. Having clarified the determination of ventilation between caves and atmosphere, it was necessary to characterize and monitor the main factors involved in soil ventilation and consequent CO2 emissions to the atmosphere. From these measurements arose the publications corresponding to chapters 3 and 4 of this thesis. In Chapter 3, is shown that in El Llano de los Juanes significant decreases occur in the CO2 molar fraction of both soil (25 cm) and borehole (7 m) when the wind is strong, as represented a high friction velocity. These decreases in the CO2 molar fraction involve CO2 emissions to the atmosphere which are detected by the Eddy Covariance tower. Also, was observed that on days with light winds (calm), CO2 accumulated in the soil, whereas during windy days large quantities of CO2 were emitted to the atmosphere. Chapter 4 presents observations from the vertical CO2 profile located in Balsa Blanca, with significant variations in the underground CO2 molar fraction in just a few hours. These variations can result in an increase or decrease of 400% compared with previous values. It is observed that changes in the molar fraction are due to changes in atmospheric pressure. There are two patterns in the subterranean CO2 fluctuations, one daily with two cycles per day due to atmospheric tides and another cycle repeating every 3-4 days (atmospheric synoptic scale) due to transitions between low and high pressure systems. Also we observed that synoptic-scale pressure changes can affect the CO2 fluxes detected by an eddy covariance tower. On days with rising pressure, the downward CO2 flux is higher than on days with falling pressure because on these days CO2 respired by plants tends accumulates in the soil.