Composite samples of bulk rainwater were collected atop the 23 meter height sampling tower at Tudor Hill, Bermuda (https://bios.asu.edu/tudorhill/facility-description), on an approximately weekly basis from November 2018 through March 2020, bracketing the four BAIT project cruises. Rainwater samples were collected in acid-cleaned 2-Liter wide-mouth fluorinated high-density polyethylene bottles (Nalgene) using an automatic rain sampler (N-Con Systems ADS 00-120); rain samples were subsequently acidified to 0.4% (v/v) with 6 M ultrapure hydrochloric acid (Fisher Optima) in the collection bottles, and then after two months the acidified, unfiltered samples were transferred into acid-cleaned 125 mL low-density polyethylene bottles (Nalgene) for analysis of "total-dissolvable Al" (TDAl; Tian et al. 2008). Field blanks for the rainwater (125 mL ultrapure deionized water) were deployed on the Tudor Hill tower and processed in the same manner as samples, but without opening the rain sampler.
Aluminum concentrations in the acidified rainwater were determined by inductively-coupled plasma mass spectrometry (ICP-MS, Thermo Fisher Scientific ElementXR) without preconcentration, using calibration standards prepared in 0.4% ultrapure hydrochloric acid (Fisher Optima) and yttrium as an internal standard. The field blank value for rainwater was 0.015 µmol TDAl, which equates to a concentration of 0.15 µmol L-1TDAl for a typical rain sample volume of 100 mL. The limit of detection for aluminum in blank-corrected rain samples was estimated as 0.90 nmol per sample, from three times the standard deviation on the mean of replicate analyses of the single field blank (in the absence of a replicate field blank for rain). In the absence of duplicate samples for rainwater, we assume an overall uncertainty on TDAl concentrations of less than ±10% (one-sigma), based on duplicate, separate-day analyses of rain samples by ICP-MS.
