For the monthly mean SST, we used the National Oceanic and Atmospheric Administration Extended Reconstructed SST version 531. MJO and its teleconnection are also constructed from two reanalysis datasets: OLR was derived from NOAA satellite data and zonal winds were from the NCEP–DOE reanalysis32. Fire weather index (FWI) and geopotential height was obtained from European Center for Medium-Range Weather Forecasts reanalysis v533. Daily outputs, including zonal winds and FWI, were used. The analysis is conducted for all seasons and covers the same period of fire emission data (2003–2020)
Fire emission data is from the fourth generation of the Global Fire Emissions Database (GFED4, https://www.globalfiredata.org) burned area and emission data set, which provides global monthly, daily, and 3hourly emission data at 0.25° spatial resolution from 2003 to 202034.
MJO-related diagnostics. The MJO life cycle is classified by eight MJO phases based on longitudinal locations of the MJO-scale convection. To obtain MJO-phase-composited fields, we calculated daily anomalies and applied a 20-100-days bandpass filter. We also calculated two leading empirical orthogonal functions (EOFs) using the filtered outgoing longwave radiation (OLR), zonal winds at 850hPa, and 200hPa averaged over 10°S-10°N. Using EOFs and filtered OLR and 250hPa and 850hPa zonal winds, we calculated the real-time multivariable MJO (RMM) indices through the projection. The MJO phase 2 and 3 correspond to strong convection over the eastern Indian Ocean, and MJO phase 5 and 6 is characterized by deep convection over the western Pacific. We only select days related to MJO when MJO magnitude is greater than 1. Both concurrent and lag composites were obtained. The Student’s t-tests were used to identify where the MJO-phase-composited anomaly is significantly different from zero.
Data related to this paper can be downloaded from the following:
NOAA/NCEP reanalysis, https://www.esrl.noaa.gov/psd/data/reanalysis/reanalysis.shtml
ECMWF reanalysis, https://www.ecmwf.int/en/forecasts/datasets/browse-reanalysis-datasets
All the data generated in this analysis will be deposited at the server clipas.soest.hawaii.edu and are available upon requests.