MET Tool: MTD

MODE-Time-Domain: General

MODE-Time-Domain Functionality

The MODE-Time-Domain (MTD) tool was added in MET version 6.0. It applies an object-based verification technique in comparing a gridded forecast to a gridded analysis. It defines 3-dimensional space/time objects, tracking 2-dimensional objects through time. It writes summary object information to ASCII statistics files and writes object fields to NetCDF format. The MTD tool can be used to quantify the duration of events and timing errors.

MODE-Time-Domain Usage

View the usage statement for MODE-Time-Domain by simply typing the following:

mtd

The forecast and observation fields must be on the same grid. You can use copygb to regrid GRIB1 files, wgrib2 to regrid GRIB2 files, or use the automated regridding functionality within the MET config files.

At a minimum, the -fcst and -obs options must be used to specify the data to be processed. Alternatively, the -single option specifies that MTD should be run on a single dataset. The -config option specifies the name of the configuration file.

cindyhg Tue, 06/25/2019 - 08:13

Configure

MTD: Configure

Start by making an output directory for MTD and changing directories:

mkdir -p ${METPLUS_TUTORIAL_DIR}/output/met_output/mtd

cd ${METPLUS_TUTORIAL_DIR}/output/met_output/mtd

The behavior of MTD is controlled by the contents of the configuration file passed to it on the command line. The default MTD configuration file may be found in the data/config/MTDConfig_default file.

Prior to modifying the configuration file, make a copy of the default:

cp ${MET_BUILD_BASE}/share/met/config/MTDConfig_default MTDConfig_tutorial

The configuration items for MTD are used to specify how the space-time-object-based verification approach is to be performed. Just as MODE may be used to compare any two fields, the same is true of MTD. When necessary, the items in the configuration file are specified separately for the forecast and observation fields. In most cases though, users will be comparing the same forecast and observation fields. The configurable items include specifications for the following:

The verification domain.
The forecast and observation fields and vertical levels or accumulation intervals to be compared.
The forecast and observation object definition parameters.
Options to filter out objects that don't meet a minimum volume.
Matching/merging weights and interest functions.
Total interest threshold for matching/merging.
Flags to control output files.

For this tutorial, we'll configure MTD to process the same series of data we ran through the Series-Analysis tool. Just like MODE, MTD compares a single forecast field to a single observation field in each run.

Open up the MTDConfig_tutorial file for editing with the text editor of your choice and edit it as follows:

vi MTDConfig_tutorial

Set the fcst dictionary as follows:

fcst = {
   field = {
      name = "APCP";
      level = "A03";
   }
conv_radius = 2;
   conv_thresh = >=2.54;
}

Set the obs dictionary as follows:

obs = {
   field = {
      name = "APCP_03";
      level = "(*,*)";
   }
   conv_radius = 2;
   conv_thresh = >=2.54;
}

Set the minimum volume for MET evaluation to 0:

min_volume = 0;

This retains all objects regardless of their calculated volume.

Save and close the configuration file.

cindyhg Tue, 06/25/2019 - 08:15

Run

MTD: Run

First, we need to prepare our observations by putting 1-hourly StageII precipitation forecasts into 3-hourly buckets. Create an output directory:

mkdir -p sample_obs/ST2ml_3h

Run the following PCP-Combine commands to prepare the observations:

pcp_combine -sum 00000000_000000 01 20050807_030000 03 \

sample_obs/ST2ml_3h/sample_obs_2005080703V_03A.nc \

-pcpdir ${METPLUS_DATA}/met_test/data/sample_obs/ST2ml

pcp_combine -sum 00000000_000000 01 20050807_060000 03 \

sample_obs/ST2ml_3h/sample_obs_2005080706V_03A.nc \

-pcpdir ${METPLUS_DATA}/met_test/data/sample_obs/ST2ml

pcp_combine -sum 00000000_000000 01 20050807_090000 03 \

sample_obs/ST2ml_3h/sample_obs_2005080709V_03A.nc \

-pcpdir ${METPLUS_DATA}/met_test/data/sample_obs/ST2ml

pcp_combine -sum 00000000_000000 01 20050807_120000 03 \

sample_obs/ST2ml_3h/sample_obs_2005080712V_03A.nc \

-pcpdir ${METPLUS_DATA}/met_test/data/sample_obs/ST2ml

pcp_combine -sum 00000000_000000 01 20050807_150000 03 \

sample_obs/ST2ml_3h/sample_obs_2005080715V_03A.nc \

-pcpdir ${METPLUS_DATA}/met_test/data/sample_obs/ST2ml

pcp_combine -sum 00000000_000000 01 20050807_180000 03 \

sample_obs/ST2ml_3h/sample_obs_2005080718V_03A.nc \

-pcpdir ${METPLUS_DATA}/met_test/data/sample_obs/ST2ml

pcp_combine -sum 00000000_000000 01 20050807_210000 03 \

sample_obs/ST2ml_3h/sample_obs_2005080721V_03A.nc \

-pcpdir ${METPLUS_DATA}/met_test/data/sample_obs/ST2ml

pcp_combine -sum 00000000_000000 01 20050808_000000 03 \

sample_obs/ST2ml_3h/sample_obs_2005080800V_03A.nc \

-pcpdir ${METPLUS_DATA}/met_test/data/sample_obs/ST2ml

Rather than listing 8 input forecast and observation files on the command line, we will write them to a file list first. Since the 0-hour forecast does not contain 3-hourly accumulated precip, we will exclude that from the list. We will use the 3-hourly APCP output from PCP-Combine that we prepared above:

ls -1 ${METPLUS_DATA}/met_test/data/sample_fcst/2005080700/wrfprs* | egrep -v "_00.tm00" > fcst_file_list

ls -1 sample_obs/ST2ml_3h/sample_obs* > obs_file_list

Next, run the following MTD command:

mtd \

-fcst fcst_file_list \

-obs obs_file_list \

-config MTDConfig_tutorial \

-outdir . \

-v 2

Just as with MODE, MTD applies a convolution operation to smooth the data. However, there are two important differences. In MODE, the convolution shape is a circle (radius = conv_radius). In MTD, the convolution shape is a square (width = 2*conv_radius+1) and for time t, the values in that square are averaged for times t-1, t, and t+1. Convolving in space plus time enables MTD to identify more continuous space-time objects.

If your data has high enough time frequency that the features at one timestep overlap those at the next timestep, it may be well-suited for MTD.

cindyhg Tue, 06/25/2019 - 08:16

Output

MTD: Output

The MTD output typically consists of 6 files: 5 ASCII statistics files and 1 NetCDF object file. MTD does not create any graphical output. In this example, the output is written to the current mtd directory as we requested on the command line.

mtd_20050807_030000V_2d.txt
mtd_20050807_030000V_3d_pair_cluster.txt
mtd_20050807_030000V_3d_pair_simple.txt
mtd_20050807_030000V_3d_single_cluster.txt
mtd_20050807_030000V_3d_single_simple.txt
mtd_20050807_030000V_obj.nc

The MTD output file naming convention begins with mtd_ followed by the last valid time it encountered. The output file names may be modified using the output_prefix option in the configuration file, which should be used to prevent the output of one run from over-writing the output of a previous run. The 6 MTD output files are described briefly below:

The NetCDF object file ends in .nc and contains gridded fields of the raw data, simple object indices, and cluster object indices for each forecast and observed timestep.
The ASCII file ending with _2D.txt contains many columns similar to the output of MODE. This data summarizes the 2-dimensional object attributes for each individual time slice of the 3D forecast and observation objects.
The ASCII files ending with _single_simple.txt and _single_cluster.txt contain 3D space-time attributes for simple and cluster objects, respectively.
The ASCII files ending with _pair_simple.txt and _pair_cluster.txt contain 3D space-time attributes for pairs of simple and cluster objects, respectively.

Use the ncview utility to view the NetCDF object output of MTD:

ncview mtd_20050807_030000V_obj.nc &

Select the variable named fcst_raw and click the time index to advance through the timesteps. Now, do the same for the fcst_object_id variable.

Notice that the objects are defined in the active areas in the raw fields. Also notice some features merging (i.e. combining) as time passes while other features split (i.e. break apart). While they may be disconnected at a particular timestep, they remain part of the same space-time object.

Next, explore the ASCII output files and pay close attention to the header columns.

Notice the generalization of the 2D MODE object attributes to 3 dimensions. Area measure becomes volume. MTD measures the object speed. Each object has a beginning and ending time.

cindyhg Tue, 06/25/2019 - 08:17