NOAA/WDS Paleoclimatology - Monsoon Asia Drought Atlas (MADA)

STUDY NOTES: Monsoon Asia Drought Atlas (MADA). Palmer Drought Severity Index reconstructions based on a network of Asian Tree ring data. Data files and their contents are as follows: 1) jja-mada.txt - 534 grid point reconstructions of JJA PDSI for the Asian monsoon region. These are the "Monsoon Asia Drought Atlas" (MADA) reconstructions from tree rings that are published in Cook et al. Science 2010. There are 534 grid point reconstructions on a 2.5x2.5° grid. The data are in column-ASCII form, one reconstruction per column and the grid point number is at the top of each column. Column 1 has the years and the total time period covered is AD 1300-2005. Missing values are indicated by -99.999. These reconstructions are actually ensemble averages of 24 separate reconstructions based on four search radii (500, 1000, 2000, and 3000 km) and six correlation weightings. See the Cook et al. 2010 Science paper SOM for details. The tree-ring estimates of JJA PDSI actually end in 1989 and the instrumental data from 1990 to 2005 have been appended to the ends of the reconstructions to bring them up to date. To do this in as seamless a manner as possible, the reconstructions have been scaled to have the same means and standard deviations as the actual data over the 1951-1989 calibration period. 2) jja-mada-xy.txt - 534 longitude-latitude pairs that locate the grid point reconstructions. The lon-lat pairs are ordered to agree with the column number above each reconstruction. 3) JJA_PDSI_Act.txt - 534 actual JJA PDSI series used for calibration and verification. These series are organized columnwise in exactly the same way as the reconstructions and each column number heading matches the number of the reconstruction. The data cover the 1870-2005 period of available data from the Dai-Trenberth-Qian. The original data are available at http://www.cgd.ucar.edu/cas/catalog/climind/pdsi.html. However, the data provided here have been modified to fill in missing values with weighted estimates from immediately adjacent grid points using a 9-point "Queen's Case" adjacency model. This has also resulted in a small degree of local spatial smoothing to the data. So there is not one-to-one agreement between what is provided here and what is directly available from the Dai-Trenberth-Qian website. See the Science paper SOM for details. 4) jja-mada-calver.txt - calibration and verification statistics for each of the 534 JJA PDSI reconstructions, with longitude- latitude coordinates for their locations. Five statistics are reported for each reconstruction: CRSQ (calibration period R2), CVRE (calibration period leave-one-out cross-validation reduction of error - essentially the same as Allen's PRESS), VRSQ (verification period square of the Pearson correlation coefficient), VRE (verification period reduction of error), and VCE (verification period coefficient of efficiency). The calibration period is 1951-1989 and the verification period is 1920-1950. See the Science paper SOM for why these periods were chosen and also why regions of weak verification may be as much due to poor instrumental data quality as due to poor estimation by the tree rings.
ABSTRACT SUPPLIED BY ORIGINATOR: The Asian monsoon system affects more than half of humanity worldwide, yet the dynamical processes that govern its complex spatiotemporal variability are not sufficiently understood to model and predict its behavior, due in part to inadequate long-term climate observations. Here we present the Monsoon Asia Drought Atlas (MADA), a seasonally resolved gridded spatial reconstruction of Asian monsoon drought and pluvials over the past millennium, derived from a network of tree-ring chronologies. MADA provides the spatiotemporal details of known historic monsoon failures and reveals the occurrence, severity, and fingerprint of previously unknown monsoon megadroughts and their close linkages to large-scale patterns of tropical Indo-Pacific sea surface temperatures. MADA thus provides a long-term context for recent monsoon variability that is critically needed for climate modeling, prediction, and attribution.