JRA-25 Atlas > Methodology
|
JRA-25 Atlas > Methodology |
This Atlas contains maps, cross sections and plots of time series. Maps and cross sections include annual, seasonal and monthly averages. Figures for interannual variability show standard deviations for the 26 years that are calculated and drawn for annual averages. Seasonal averages are shown for Dec. - Feb., Mar. - May, Jun. - Aug. and Sep. - Nov. except for figures relating to tropical cyclones. Most of the figures are drawn directly from monthly averages or monthly normal values available from the JRA-25 official web page. For the figures for which additional processing was carried out, the methodologies are outlined below.
Most figures are drawn using T106 Gaussian grid data, but some figures are from 2.5-degree latitudinal-longitudinal grid data.
These elements are calculated from daily data. Six-hour forecasts from 00, 06, 12 and 18 UTC in the same day are used to derive the daily data.
The maximum (minimum) value from the four forecast values of maximum (minimum) temperature is picked up as the daily maximum (minimum) temperature. The figures of "Daily maximum (minimum) temperature" are drawn as monthly averages of the daily maximum (minimum) temperatures. Figures of "Daily maximum wind" are calculated and drawn in the same way.
Daily precipitation is calculated as is done for daily maximum temperature. Four forecast values of precipitation are summed up into a daily precipitation. The monthly maximum values of the daily precipitation amounts are produced for 312 months (12 months x 26 years), and then monthly normal values (26 year average) are calculated and drawn.
A standard deviation of wind speed is derived from a square root of the sum of the variances (squared standard deviations) of the zonal and meridional wind components except for the daily maximum wind. Since the daily maximum wind is a scalar variable, a standard deviation is calculated from its magnitude.
The figures are drawn as the rate of monthly precipitation to GPCP ver.2 (Adler et al. 2003). The areas where the GPCP precipitation is less than 0.1mm/day are not colored because the rate may become extremely large for such a small GPCP value.
Tropical cyclones are defined by referring to Hart (2003).
Figures of "Number of tropical cyclone days" (unit: day/month) are drawn in accordance with the "tropical cyclone" as described above. Here, each season is defined as Jan. - Mar., Apr. - Jun., Jul. - Sep. and Oct. - Dec., considering seasonal changes in tropical cyclone activities.
Time series of the accumulated number of tropical cyclone days are also drawn in accordance with tropical cyclones as defined above. The values are the sums of tropical cyclone days for each month for each basin. Observational values calculated from best track data of tropical cyclones are also drawn as a reference. Figures in this Atlas are not the same as figures in Hatsushika et al. (2006). The latter indicates how many tropical cyclones in best track data appear in JRA-25 data.
Q1 is an apparent heat source and Q2 is an apparent moisture sink, which are introduced by Yanai et al. (1973). In this Atlas, Q1 and Q2 are derived from vertical integration of monthly average values of 6-hour forecast data from 1000hPa to 100hPa. Q1 is calculated by summing the heating rate of convective condensation, large scale condensation, longwave radiation, solar radiation and vertical diffusion. Q2 is a negative value calculated by summing the moistening rate of convection, large scale condensation and vertical diffusion.
A stream function is an indicator for non-divergent winds. In an area with an anticyclonic circulation, the stream function is positive (negative) in the northern (southern) hemisphere. On the other hand, a velocity potential indicates a large scale divergence or convergence. The maximum (minimum) portions of positive (negative) velocity potential indicate centers of large scale convergence (divergence). Figures of stream functions and velocity potentials are prepared for the levels where divergent and convergent motions are clearly seen; i.e., isobaric surfaces of 200hPa and 850hPa, and isentropic surfaces of 350K and 300K. Standard deviations are not calculated for the 300K surface which often intersects with the ground.
JMA uses this variable to monitor the activities of synoptic scale disturbances (travelling anticyclones and cyclones). It is defined as the mean of the squares of zonal and meridional wind speeds at 500hPa surface in 6-hourly analysis which are 2-8-day band-pass-filtered after Duchon (1979).
Solar radiation is an energy source of the atmospheric and the oceanic circulation. For the globe as a whole, outgoing longwave radiation in accordance with Earthfs surface temperature balances with incoming solar radiation. However, incoming radiation exceeds outgoing radiation in the tropics, while outgoing radiation exceeds incoming radiation in the polar regions. Heat and momentum flow from the tropics toward the polar regions compensate this geographical imbalance.
JMA uses "Northward flux of horizontal momentum in the upper troposphere" and "Northward heat flux at 850hPa" to monitor such meridional transfer. The calculation methods follow Newell et al. (1972). They defined the component of the mean meridional circulation as fluxes by monthly zonal mean, the component of the stationary disturbances as fluxes by monthly non-zonal mean and the component of the high-frequency disturbances as fluxes by deviation from the monthly mean field derived from 6-hourly analysis data. The northward flux of horizontal momentum is calculated by integrating the values from 500hPa to 100hPa. These values are negative when momentum or heat flows toward south.
The Madden-Julian Oscillation (MJO) is an intraseasonal oscillation in which a region of active convection moves cyclically in the tropics with a period of 30 - 60 days. In order to remove short time scale variability, data from 1980 to 2003 are used to define empirical orthogonal functions (EOFs) after applying a 120-day running mean. We have partially changed MJO index modified by Wheeler and Hendon (2004) as follows:
The indices RMM1 and RMM2 are defined as scores of two principal component time series obtained by projecting data onto the multiple-variable EOFs, and root mean square of sum of the two indices are drawn.
An El Nino event is characterized by warmer SSTs than normal in the eastern equatorial Pacific, and persists typically for a half to one and a half years. The oceanic condition of cooler SSTs than normal in the same region is referred to as a La Nina event. On the other hand, it has been well known since the early 20th century that there is a see-saw in the sea level pressure between the western and eastern equatorial Pacific. This phenomenon has been called the Southern Oscillation. At present, the El Nino/La Nina phenomenon and the Southern Oscillation are recognized as different aspects of a single phenomenon. This phenomenon is often called "El Nino and the Southern Oscillation" or "ENSO" in short.
The Southern Oscillation Index (SOI) is an index to monitor ENSO defined as a difference in monthly mean sea level pressure anomalies between Tahiti, French Polynesia and Darwin, Australia. In this Atlas, the SOI is calculated by regarding the sea level pressure at the grid point of 150.0W, 17.5S as that in Tahiti and the grid point of 130.0E, 12.5S as that in Darwin. The normalized SOI is derived by dividing by the standard deviation of SOI. The SOI here is a difference in pressure anomaly divided by the standard deviation between Tahiti and Darwin. The calculation method for the normalized SOI is the same as that for the "SOI" used operationally in JMA to monitor ENSO (hereafter, operational SOI). It should be noted that the normalized SOI in this Atlas is not the same as the operational SOI, because the latter is calculated by using observational data at each point, and climatological normal and standard deviation for the latter are calculated from data of 1971 - 2000.
To draw figures, we used the Grid Analysis and Display System (GrADS) developed in the Center for Ocean-Land-Atmosphere Studies (COLA) / the Institute of Global Environment and Society (IGES). Much valuable advice concerning this Atlas was given by Japanese climate researchers and by staff of the Central Research Institute of Electric Power Industry and of the Japan Meteorological Agency. We greatly appreciate all of them.
