The P3 experiment consists of three permanent treatment plots situated on the south-facing hillslope above Many Glaciers Pond in Taylor Valley. Each plot is 15 m by 7.5 m. A trench (20 cm deep), located at the top of each plot, is used for experimental wetting events.

• Press (H): Receives yearly water additions sufficient to raise soil water to 10% water content (as observed during the melt event in Jan. 2002). This is achieved by pumping water from Many Glaciers Pond to a holding tank above the experimental plots, from which water is applied by gravity feed to a trench at the top of the hillslope during the warmest week of the year (typically the 1st or 2nd week of Jan.). The water from the trench then seeps down through the soil.
• Pulse (L): Same as above, but water additions occur in alternate years to simulate a lower frequency of thawing permafrost.
• Control (C): Trenched and instrumented, but receives no water addition treatments.

In January 2012, each treatment plot was instrumented with a network of soil moisture and temperature sensors, which are located down the center of each plot at 13 locations (blocks) at varying distances (m) from the trench and at varying depths in the soil (cm). There are 2-3 sensors deployed in each block. One is shallow, at 5 cm deep in the soil, whereas the other 1-2 sensors are deeper, >10 cm deep in the soil. The depth of the deeper sensors was dependent on how frozen the ground was at the time of deployment. Data from each plot are collected continuously on a 5 to 60-minute interval by an associated Campbell Scientific CR1000 datalogger paired with an AM1632 and AM25T multiplexer. 

Decagon 5TM soil moisture and temperature sensors are used to measure soil volumetric water content (VWC) and soil temperature in blocks 1, 2, 3, 5, 7, 9, 11 (except at 26 cm in the control plot), and 15 (except at 35, 26, and 21 cm in the control, press, and pulse plots respectively), while thermocouples measure temperature in the other blocks (4, 6, 8, 10, 11 (at 26 cm in the control plot only), 13, and 15 (at 35, 26, and 21 cm in the control, press, and pulse plots respectively). The data collection frequency has changed several times over the lifespan of the project, ranging 5-min to 1-hour collection intervals. Generally, the frequency of data collection occurs every 5 minutes during the period of wetting, then is decreased to either every 30 or 60 minutes depending on the time of year. 

Note that datalogger memory problems and sensor malfunctions over the years have resulted in some missing data. Also, due to the difficulty in calibrating the EC-TM Soil Moisture sensors to polar desert soils, the VWC data must be looked at as relative to the background for each probe.