Thank you for your patience and interest in the TIMPS version 2 dataset. The new version uses the IMERG version 7 satellite precipitation data to track mesoscale convective systems (MCSs) in the global tropics (30N to 30S). The data is available for 13 years from 2011 to 2023. In addition to using IMERG version 7, the new version includes following changes. The effect of these changes is minor and didn’t affect qualitative conclusion . I verified my published plots with new TIMPS version, and they look good. New changes =========== (1) MCS size criteria is changed from 30 IMERG pixels to 3000 km^2 (2) Convective rain rate threshold is now inclusive. In the previous version, the convective rain was defined as precipitation rate > 10 mm/hr, but in the current version it is defined as precipitation rate >= 10 mm/hr. This has minor effect on following properties: area10, vrr10, pieces10, meanrr10, medrr10, stdevrr10, and skewrr10. (3) Similar to above, in the previous version MCS properties such as area1, vrr1, pieces1, meanrr1, medrr1, stdevrr1, and skewrr1 are computed from pixels with precipitation rate > 1 mm/hr, within an MCS. In the current version they are computed from all pixels with precipitation rate >= 1 mm/hr. (4) Since tropical cyclone data is available only up to 2022-11-01, the global attribute "within_TC” is set to ""TC data not Available" after this date (5) Propagation direction is now meteorological direction (degrees from north, in clockwise direction). In the previous version it was vector direction (6) latitude, longitude, and rain rates of all pixels inside an MCS object is not stored with “TIMPS_files”, but can be accessed from “fields2D_precip_n_mcsid” data Filename and format are backward compatible: ===================================== The netcdf data filenames, variables, and their data type are same as the previous version. So, the data is backward compatible with programs that used TIMPS version 1 dataset Directory structure ================ TIMPS_files – Each netcdf file represent an MCS. The monthly directory has all MCSs that initiated in that month. MCSs that initiated towards the end of a month may live into next month but its data is stored under the month it initiated. fields2D_precip_n_mcsid – Each netcdf files has IMERG precipitation on lat/lon grid , and tracked object id on lat/lon grid. This helps map the object id to IMERG precipitation, for further analysis. lifetime_prop – This and “instant_prop” are like indices that help find MCSs with certain characteristics. MCS properties are stored individually to make the search faster. For example, the file “lifetime.nc” has duration of all MCSs that occurred in that month. So, If you want to find MCSs that lived longer than 24 hours, you can use this file to find those MCSs, quickly. All property files are ordered by id, so TIMPS id of an MCS can be obtained from “4Dobjid.nc” at the same array index with corresponding value for lifetime. Similar search can be done for through other properties, individually or in combination. instant_prop – Similar to “lifetime_prop”, this helps to search MCSs with specific property values. Downloading data: =============== The data is compressed and archived into monthly files for easy download. The following “wget” commands will help download files. The URL determines how much of data you want to download. All data can be accessed from https://home.chpc.utah.edu/~mani/TIMPS_v2/ To download TIMPS_files for 2011 wget -r -np -e robots=off -nH --cut-dirs=1 -R "index.html*" https://home.chpc.utah.edu/~mani/TIMPS_v2/TIMPS_files/2011/ If you want to download TIMPS files for all thirteen years wget -r -np -e robots=off -nH --cut-dirs=1 -R "index.html*" https://home.chpc.utah.edu/~mani/TIMPS_v2/TIMPS_files/ To download the fields 2D data for 2011 with IMERG precipitation and tracked obj id wget -r -np -e robots=off -nH --cut-dirs=1 -R "index.html*" https://home.chpc.utah.edu/~mani/TIMPS_v2/fields2D_precip_n_mcsid/2011/ Uncompressing data ================= All directories have a script uncompress_data.sh that will uncompress the monthly files. You can execute it from command line as “nohup ./uncompress_data.sh &> uc.log &”