2.3. Isoprenoid emission
Leaf-level gas fluxes were measured on site from one leaf per tree and repeated in the same tree across seasons. Measurements were made using a commercial, portable gas exchange system with infrared gas analyzer (IRGA), the LI6400XT (LiCor, U.S.A.). A hydrocarbon filter (Restek Pure Chromatography, Restek Corporations, U.S.A.) was installed at the air inlet of the IRGA to remove hydrocarbons from incoming ambient air. All tubing in contact with the sampling air was PTFE (a material that is non-reactive to hydrocarbons). Before each measurement, a blank was sampled from the empty leaf chamber (see more details on blank samples in Table S1, Supporting Information). One leaf was separately enclosed in the leaf chamber with standard conditions — photosynthetic photon flux density (PPFD) at 1000 μmol m-2s-1 and leaf temperature at 30ºC — until net assimilation (An), stomatal conductance (gs) and internal CO2 concentration (Ci) were stable. The stability criterion for measurements was assigned as up to one standard deviation of the mean An. The flow rate of air going into the leaf chamber was 400 μmol s-1, CO2 and H2O concentrations were 400 μmol mol-1and 21 mmol mol-1 (relative humidity of ~ 60%), respectively. Air exiting the IRGA leaf chamber was routed to flow through adsorbent cartridges (stainless silico steel tubes filled with Tenax TA and Carbograph 5 TD adsorbents) at a rate of 200 sccm for 10 min, which resulted in 2 L air samples for isoprenoid chemical analysis. The isoprenoids accumulated in the adsorbent cartridges were determined subsequently by laboratory analysis. Samples from December 2018 and May 2019 were analysed in the State University of Amazonas (UEA, Brazil), and samples from September 2019 were analysed in the Max Planck Institute for Chemistry (MPIC, Germany).
In UEA, cartridges were analyzed with a thermal desorption system (TD; Markes International, U.K.) interfaced with gas chromatograph-mass spectrometer and flame ionization detectors (GC-MS-FID; 7890B-GC and 5977A-MSD series, Agilent Technologies, U.S.A.). The cartridges were loaded in the TD automatic sampler (TD-100, Markes International, U.K.), which connects to the thermal desorption system. Then, samples were dried by purging for 5 min with 50 sccm of ultrahigh-purity helium (all flow vented out of the split vent) and transferred (300ºC for 10 min with 50 sccm of ultrapure nitrogen) to the thermal desorption cold trap held at -10ºC (Unity Series 1, Markes International, U.K.). During GC injection, the trap was heated to 300ºC for 3 min while backflushing with carrier gas (helium) at a flow rate of 6.0 sccm directed into the column (Agilent HP-5, 5% phenyl methyl siloxane capillary, 30.0 m x 320 μm x 0.25 μm). The oven ramp temperature was programmed with an initial hold of 6 min at 27ºC, followed by an increase to 85ºC at 6ºC min-1, followed by a hold at 200 ºC for 6 min. We confirmed the identification of emitted isoprenoids from the samples by comparison of retention time with a solution of authentic liquid standards in methanol (Sigma-Aldrich, U.S.A.) and comparison to the library of the National Institute of Standards and Technology (NIST). The GC-MS-FID was calibrated at least three times before the analysis of the sample cartridges; with calibration curves that were generated by injecting different amounts of gas standard (27 biogenic VOCs gas mixture, by Apel & Riemer Environmental Inc., U.S.A.) into separate cartridges, a mean correlation coefficient >= 0.98 was obtained, and the LOD quantified as 27 pptv.
In MPIC, the cartridges were analyzed through thermal-desorption gas chromatography time of flight mass spectrometry (TD-GC-TOF-MS, Bench ToF Tandem Ionisation from Markes International, U.K.). The analysis consisted of three main steps: desorption of the analytes from the cartridges (TD), separation of the analytes through gas chromatography (GC), and quantification and identification of the analytes through time-of-flight mass spectrometry (ToF-MS). Samples were first dried by purging them for 5 min with a flow of ultrapure N2 at 50 ml min-1, then transferred to the thermal-desorption unit. Thermal-desorption was carried out in two stages - tube desorption and trap desorption, both performed at 250°C for 10 minutes through a TD 100xr (Markes International, U.K.). The desorbed components were carried in a flow of Helium into the GC column (dimethyl T.B.S. β-cyclodextrin 0.15µm, 0.15mm ID, 25m L, from MEGA, Italy). The temperature ramp consisted of an initial 5 min at 40°C, after which the temperature was increased at a rate of 1.5 °C/min from 40°C to 150°C, and further increased at a rate of 30°C/min from 150°C to 200°C. After the GC column, the analytes were fragmented through electron impact ionization at -70 eV in the ToF. Identification was obtained by comparing the MS spectra with the MS NIST library for the same ionization energy and by injection of gas mixtures (162 VOCs gas mixture and 25 biogenic VOCs gas mixture, by Apel & Riemer Environmental Inc., U.S.A.) and liquid standards. The obtained chromatograms were integrated with TOF-DS (Markes International, U.K.). Gas standard cartridges were used to calibrate the instrument, determine the precision and LOD of the analysis, which was quantified as 23% and ~1 pptv, respectively. More information on the material and method used can be found in Zannoni et al. (2020).
For the final flux calculation, isoprenoid concentrations were determined using the sample volume that was passed through each cartridge. This volume is the integration of the mass flow rate measured and controlled by the pump used to suck the air coming out from the IRGA leaf chamber. Once the volume mixing ratios of isoprenoids (ppbv) were obtained, leaf emission fluxes were determined using the equation (F = R ppbv × Q/A), where F (nmol m-2s-1) is leaf flux of isoprenoid emission; Rppbv(nmol mol-1) is isoprenoid concentration of the sample; Q is flow rate of air into the leaf chamber (400 x 10-6 mol s-1); and A is the area of leaf within the chamber (0.06 m²). In order to calculate isoprenoid emission on a mass basis, we measured Leaf Mass per Area (LMA). LMA was calculated as the ratio of leaf dry weight to leaf area. We did not include petioles in the LMA calculation since they can be quite large for rainforest species and are usually more related to leaf positioning rather than biomass efficiency (Poorter, Castilho, Schietti, Oliveira & Costa 2018). With LMA, isoprenoid emissions were then calculated to μgC g-1 h-1.