Temperature sensitivities of MeOH and AA emissions and AA/MeOH emission ratios from physiologically active trees, detached stems and leaves, hydrated AIR, and whole ecosystems
To better understand the role of temperature in potentially enhancing AA/MeOH emission ratios, the air temperature sensitivities of MeOH and AA emission rates were characterized from branches of well-watered poplar trees, detached stems and leaves, and whole ecosystems. Well-watered poplar trees were individually placed in a growth chamber with diurnally changing air temperature (Figure 5 ). At night in the dark (20:00-6:00), considerable branch transpiration was observed together with relatively high MeOH emissions, and low to undetectable AA emissions. Under constant daytime (6:00-20:00) light conditions, net positive CO2 assimilation occurred. As observed at the leaf level, branch transpiration together with MeOH and AA emissions were strongly coupled to the diurnal pattern of air temperature, reaching maximum fluxes during the early afternoon peak in air temperature of 27 ºC at 14:00. MeOH emissions were greater than AA emissions at all air temperatures by roughly a factor of 10, except for 1 hour following light to dark transitions where a short burst in AA emissions were observed. Outside of this light-dark period, branch AA/MeOH emission ratios remained low and increased as a function of air temperature from 0-12% (Figure 5 ). A high temperature sensitivity of MeOH and AA emissions from detached poplar branch segments were also observed. Emissions of MeOH and AA and water loss from detached stem segments in the dark increased with temperature from 30-50 °C (Supplementary Figure S7 ). Similar to physiologically active leaves and branches, emissions were dominated by MeOH, resulting in AA/MeOH emission ratios below 12%. Similar temperature sensitivities of MeOH and AA emissions were also obtained from hydrated whole leaf cell wall preparations (alcohol insoluble residue, AIR). Gas-exchange analysis under controlled temperature with hydrated AIR in porous Teflon tubes showed rapid equilibration of MeOH and AA emissions at each chamber temperature in the dark (Figure 6 ). MeOH and AA steady state emissions from hydrated AIR samples increased as a function of temperature from 30-50 ºC and were completely dependent on the presence of liquid water interacting with AIR. Similar to physiologically active leaves (Figure 4 ), branches (Figure 5 ), and detached stems (Supplementary Figure S7 ), emissions from hydrated AIR were dominated by MeOH with AA/MeOH emission ratios increasing slightly with temperature but remaining below 30%.
In contrast to AA/MeOH emission ratios from physiologically active leaves, branches, detached stems, and hydrated leaf AIR samples which remained less than 30%, AA/MeOH emission ratios from drought stressed poplar branches were high ranging from 400-3,000% (Figures 2-3, Supplementary Figures S2-S5 ). Similarly, detached poplar leaves placed into the temperature-controlled chamber in the dark in a dry air stream, showed a similar pattern of suppressed MeOH emissions together with temperature stimulated emissions of the fermentation volatiles acetaldehyde, ethanol, acetic acid, and acetone (Figure 7 ). Acetaldehyde emissions peaked at 42.5 ºC, AA emissions peaked at 47.5 ºC, and the AA/MeOH emission ratio reached a maximum of 2,500% at 45 ºC.
We extended our analysis of the temperature sensitivities of AA and MeOH leaf emissions at the ecosystem scale at the Lochristi poplar plantation in Belgium and a mixed hardwood forest in Alabama. Likely due to the very low ambient temperatures at night in Belgium during the growing season (averaging around 15 °C), no significant night time AA concentration or vertical flux was observed (Figure 8a,b) . In contrast, night-time MeOH emissions of 10 µmol ha-1day-1 and ambient concentrations of 0.5 ppb were detected. Similar to emissions from physiologically active poplar leaves (Figure 4 ) and branches during the day (Figure 5 ), average ecosystem emission rates of AA and MeOH during the 2015 growing season showed a clear diurnal pattern reaching maximum values in the afternoon together with air temperature. AA emissions and ambient concentrations closely followed diurnal increases in air temperature with both reaching a maximum in the afternoon (16:30). In contrast, MeOH emissions peaked around midday (12:00), and this pattern resulted in the AA/MeOH emission and concentration ratios peaking in the early evening (17:30). Ecosystem AA/MeOH emission ratios increased as a function of air temperature from a low of 28% in the early morning to 158% in the afternoon (Figure 8b ). Likewise, AA/MeOH concentration ratios also increased as a function of air temperature from a low value of 41% in the early morning to 188% in the afternoon (Figure 8a ).
We also analyzed ambient concentration time series of AA and MeOH as a function of ambient temperature above a mixed forest canopy in Alabama, USA and a citrus grove in California. Similar to the poplar plantation in Belgium, ambient concentrations of AA and MeOH above the mixed forested canopy in Alabama and citrus grove in California followed strong diurnal patterns tightly coupled with air temperature, with air temperature having a positive impact on AA/meOH ambient concentration ratios (Figure 9 ). While ambient concentrations increased during the day relative to the night in the Alabama site (Figure 9a ) in a similar way to the Belgium site (Figure 8a ), ambient concentrations decreased during the day at the California site (Figure 9b ). This may be related to a shallow boundary layer at night where emissions are concentrated, followed by a rapid increase in the boundary layer height during the day due to increased turbulent mixing. In addition, the site had the highest ambient concentrations of AA (3.6-5.6 ppb) when compared with the Belgium and Alabama sites. Despite the high ambient backgrounds and dilution effect due to turbulent mixing during the day, the MeOH/AA concentration ratio at the California grove increased linearly with air temperature from a low of 19% at night to 29% in the afternoon. This is comparable to the pattern at the Alabama mixed hardwood forest site where MeOH/AA concentration ratios also increased linearly with air temperature from < 5% during early mornings below 20 ºC, to high values of 25% during afternoon air temperatures up to 32 ºC. Given that the MeOH/AA concentration ratios increased with temperature, but remained below 30%, these values are comparable to AA/MeOH temperature sensitivities observed in the hydrated AIR, leaf, and branch poplar studies.