See the associated publication: Steiger, N. J., Smerdon, J. E., Cook, E. R., and Cook, B. I., A reconstruction of global hydroclimate and dynamical variables over the Common Era, Scientific Data, 5:180086, doi: 10.1038/sdata.2018.86. This database provides the first truly global reconstructions of hydroclimate along with associated climate dynamical variables over the past two thousand years. The reconstructions were made with an improved data assimilation reconstruction approach that optimally combines 2,978 paleoclimate proxy-data time series with the physical constraints of an atmosphere--ocean climate model. Three separate global reconstructions were created from the years 1 to 2000 CE, targeting annual means (defined as April to the next calendar year March), the boreal growing season of June, July, and August (JJA), and the austral growing season of December, January, and February (DJF); each of these reconstructions is contained in a separate NetCDF file. The reconstructions include three global variables gridded at about 2 degree resolution: surface temperature at 2 m, the Palmer drought severity index (PDSI), and the standardized precipitation evapotranspiration index (SPEI). Dynamical variables also included for each reconstruction: the global mean temperature, the North Atlantic sea surface temperature index which is the non-detrended and non-smoothed version of the Atlantic multidecadal oscillation (AMO), the monthly Nino SST indices (Nino 1+2, 3, 3.4, 4), the monthly equatorial Pacific zonal SST gradient, and the location of the intertropical convergence zone (ITCZ) in 11 longitudinal zones. A 100 member ensemble (randomly drawn from the original reconstruction ensemble) is available externally due to Zenodo file size restrictions: http://clifford.ldeo.columbia.edu/nsteiger/recon_output/phyda_ens/ Please cite the associated publication if you use this data set. Contact the author Nathan Steiger (http://www.ldeo.columbia.edu/~nsteiger/contact.html) for more information about the details of the reconstructions.

Version 2 update: Fixed incorrect percentile values for index variables.