Soil was collected from two sites in the Sonoran Desert: one inside the city of Phoenix, Arizona, at Piestewa Peak, and the other from outside the city, at White Tank Regional Park. Soil was collected in bulk from the top 10 cm from experimental plots that have been receiving experimental N fertilization as 60 kg N ha−1 yr−1, as well as control plots outside of the fertilization study. Soils were frozen until ready for the experimental incubation. To prepare for a laboratory incubation, soil was sieved to 2 mm and 25±0.5 grams were placed into 40 mL clear glass vials with a screw-on lid fitted with gas-impermeable septa. Deionized water was added to simulate precipitation as 5 mm pulse events (average summer size for Phoenix over the past 5 years) or 7.5 mm pulse events (50% increase) either every 2 weeks (simulating the average number of summer events occurring over the past 5 years, spread evenly) or every 4 weeks (reduced frequency). Incubation vessels were placed into an incubator at 30.5 degrees C for 8 weeks to simulate average Phoenix, Arizona temperatures. Vessels were loosely capped to allow for gas exchange. Full-spectrum light was added to the incubator on a timer for daylight (6 a.m.-6 p.m.). This procedure was repeated on a separate set of vials incubated at 32.5 degrees C to simulate a 2-degree C increase from future climate change. Therefore, our treatments consisted of a fully-factorial design manipulating four global change factors: urbanization (2 levels: urban, exurban), N fertilization (2 levels: control vs. fertilized), precipitation (4 levels: 5 mm and 7.5 mm pulse sizes at both every 2 and 4 weeks), and temperature (2 levels: 30.5 and 32.5˚C). Each treatment combination was replicated four times, yielding 128 separate incubation vials.
CO2 flux (soil respiration) was measured in incubation vessels using a LI-COR Infrared Gas Analyzer (LI-7000, LI-COR, Lincoln, NE) regularly for the duration of the eight-week period. Vessels were capped tightly and flushed with CO2-free air for 45 s at 5 psi. Sealed vessels were placed back in the incubator for 5-8 hours. 2 mL of headspace gas, including respired CO2, from each vessel was removed with a syringe and injected into the LI-COR Gas Analyzer. To calculate CO2 concentration, the maximum integration value (area under the curve) was recorded after every sample injection and compared against a standard curve. Net C mineralization was then calculated from the CO2 concentration using the equation:
> (AWc\*[CO2] \* V)/(R \* T)/(S \* d)
Where AWc is the atomic weight of C (12 g mol-1), V is the headspace volume, R is the gas constant (0.082 L atm K-1 mol-1), T is the temperature, S is the initial dry soil mass added to the vials, and d is the days of CO2 accumulation. After measurement, vessels were placed back into the incubator with their caps loosened. This process was repeated throughout the incubation, initially every other day to capture the initial flush of C mineralization, becoming less frequent during the later stages of the incubation. After the eight-week incubation period, soil water content (SWC) was measured. 5±0.5 g of soil was taken from each incubation vessel and dried at 105 degrees C for 48 hours. To measure inorganic N (NO3 + NO2-N and NH4-N), 10±0.5 grams of fresh soil was extracted into 50 mL 2M KCl. Samples were centrifuged at 15,000 rpm for 10 minutes then frozen until ran on a flow injection analyzer (Lachat QC8000), and expressed as the amount of N per g of dry-equivalent soil.
