Monitoring and Understanding the Atmosphere in an Enclosed Environment

The Landscape Evolution Observatory (LEO) in Biosphere 2 consists of three identical 11m x 33m slopes with ~1m soil-depth (without vegetation) and about 1800 sensors each for the measurement of soil and atmospheric properties in an enclosed environment. Instrumentation includes below ground measurements of soil water potential, content and temperature; and above ground measurements of net radiation, temperature, relative humidity and wind, among several other variables. The integrated measurements at LEO offer an exceptional opportunity to gain insight about the coupling processes between soil and atmosphere within Biosphere 2. Furthermore, computer models have been developed to simulate the LEO biogeochemical and hydrological processes, and they act as a bridge to transfer our understanding from within Biosphere 2 to the real world (Biosphere 1).

The above ground instrumentation for each slope is organized in 5 retractile poles with sensors at 5 different vertical levels, totaling 24 Vaisala HMP60 sensors for air temperature and relative humidity, 24 Davis Instruments DVI7911 cup anemometer and wind vane, 2 KippZonen CNR4 net radiometer and 1 Campbell Scientific CSAT3 3D anemometer; precipitation is monitored by 1 EX81 (SeaMetrics) flow meter and data from an Automatic Weather Station outside the building are also available. Data are stored every 15 minutes and during selected rainfall periods they are stored every minute.

In the current study we present a detailed analysis of the enclosed atmosphere at LEO during the period from 1 October 2016 to 31 January 2017, which covered a PERiodic Tracer Hierarchy experiment (PERTH; 6 November – 26 December 2016). In this experiment, two 12 mm/h rainfall pulses of 3 hours each and separated by 7 hours were performed every 3.5 days in each slope. This time period covered a wide range of soil moisture states, at the beginning the soil was mostly dry with little to no variation in its water content; during PERTH a periodic steady state of the soil was reached and after PERTH a monotonically decreasing soil water content was observed.

A thorough quality control on the observations was performed, including range verification, outlier analysis, and finally a spatial and temporal consistency check. Vertical profiles of temperature and relative humidity were analyzed showing, as expected for an enclosed atmosphere, a very high stability. Surprisingly at first, temporal evolution of temperature does not show the expected negative correlation with precipitation; this was probably due to the high stability observed at LEO, which inhibited the evaporation from the bare soil below. This hypothesis is explored in more detail by means of the analysis of temporal evolution of the available variables, their anomalies and their differences respect to the outside weather station measurements.