t.rast.mapcalc - GRASS GIS manual

Note: This document is for an older version of GRASS GIS that has been discontinued. You should upgrade, and read the current manual page.

NAME

t.rast.mapcalc - Performs spatio-temporal mapcalc expressions on temporally sampled maps of space time raster datasets.

KEYWORDS

temporal, algebra, raster, time

SYNOPSIS

t.rast.mapcalc

t.rast.mapcalc --help

t.rast.mapcalc [-ns] inputs=name[,name,...] expression=string [method=name[,name,...]] output=name basename=basename [nprocs=integer] [--overwrite] [--help] [--verbose] [--quiet] [--ui]

Flags:

-n: Register Null maps
-s: Check the spatial topology of temporally related maps and process only spatially related maps
--overwrite: Allow output files to overwrite existing files
--help: Print usage summary
--verbose: Verbose module output
--quiet: Quiet module output
--ui: Force launching GUI dialog

Parameters:

inputs=name[,name,...] [required]: Name of the input space time raster datasets
expression=string [required]: Spatio-temporal mapcalc expression
method=name[,name,...]: The method to be used for sampling the input dataset; Options: start, during, overlap, contain, equal, follows, precedes; Default: equal
output=name [required]: Name of the output space time raster dataset
basename=basename [required]: Basename for output raster maps; A numerical suffix separated by an underscore will be attached to create a unique identifier
nprocs=integer: Number of r.mapcalc processes to run in parallel; Default: 1

DESCRIPTION
NOTES
EXAMPLES
- Semantic label filtering
SEE ALSO
AUTHOR

DESCRIPTION

t.rast.mapcalc performs spatio-temporal mapcalc expressions on maps of temporally sampled space time raster datasets (STRDS). Spatial and temporal operators and internal variables are available in the expression string. The description of the spatial operators, functions and internal variables is available in the r.mapcalc manual page. The temporal functions are described in detail below.

This module expects several parameters. All space time raster datasets that are referenced in the mapcalc expression must be listed in the inputs option. The first space time raster dataset that is listed as input will be used to temporally sample all other space time raster datasets. The temporal sampling method can be chosen using the method option. The order of the STRDS's in the mapcalc expression can be different to the order of the STRDS's in the input option. The resulting space time raster dataset must be specified in the output option together with the basename of generated raster maps that are registered in the resulting STRDS. Empty maps resulting from map-calculation are not registered by default. This behavior can be changed with the -n flag. The flag -s can be used to assure that only spatially related maps in the STRDS's are processed. Spatially related means that temporally related maps overlap in their spatial extent.

The module t.rast.mapcalc supports parallel processing. The option nprocs specifies the number of processes that can be started in parallel.

A mapcalc expression must be provided to process the temporal sampled maps. Temporal internal variables are available in addition to the r.mapcalc spatial operators and functions:

The supported internal variables for relative and absolute time are:

td() - This internal variable represents the size of the current sample time interval in days and fraction of days for absolute time, and in relative units in case of relative time.
start_time() - This internal variable represents the time difference between the start time of the sample space time raster dataset and the start time of the current sample interval or instance. The time is measured in days and fraction of days for absolute time, and in relative units in case of relative time.
end_time() - This internal variable represents the time difference between the start time of the sample space time raster dataset and the end time of the current sample interval. The time is measured in days and fraction of days for absolute time, and in relative units in case of relative time. The end_time() will be represented by null() in case of a time instance.

The supported internal variables for the current sample interval or instance for absolute time are:

start_doy() - Day of year (doy) from the start time [1 - 366]
start_dow() - Day of week (dow) from the start time [1 - 7], the start of the week is Monday == 1
start_year() - The year of the start time [0 - 9999]
start_month() - The month of the start time [1 - 12]
start_week() - Week of year of the start time [1 - 54]
start_day() - Day of month from the start time [1 - 31]
start_hour() - The hour of the start time [0 - 23]
start_minute() - The minute of the start time [0 - 59]
start_second() - The second of the start time [0 - 59]
end_doy() - Day of year (doy) from the end time [1 - 366]
end_dow() - Day of week (dow) from the end time [1 - 7], the start of the week is Monday == 1
end_year() - The year of the end time [0 - 9999]
end_month() - The month of the end time [1 - 12]
end_woy() - Week of year (woy) of the end time [1 - 54]
end_day() - Day of month from the start time [1 - 31]
end_hour() - The hour of the end time [0 - 23]
end_minute() - The minute of the end time [0 - 59]
end_second() - The second of the end time [0 - 59].

The end_* functions are represented by the null() internal variable in case of time instances.

NOTES

We will discuss the internal work of t.rast.mapcalc with an example. Imagine we have two STRDS as input, each one of monthly granularity. The first one A has 6 raster maps (a3 ... a8) with a temporal range from March to August. The second STRDS B has 12 raster maps (b1 ... b12) ranging from January to December. The value of the raster maps is the number of the month from their interval start time. Dataset A will be used to sample dataset B to create a dataset C. We want to add all maps with equal time stamps if the month of the start time is May or June, otherwise we multiply the maps. The command will look as follows:

t.rast.mapcalc input=A,B output=C basename=c method=equal \
    expression="if(start_month() == 5 || start_month() == 6, (A + B), (A * B))"

The resulting raster maps in dataset C can be listed with t.rast.list:

name    start_time              min     max
c_1     2001-03-01 00:00:00     9.0     9.0
c_2     2001-04-01 00:00:00     16.0    16.0
c_3     2001-05-01 00:00:00     10.0    10.0
c_4     2001-06-01 00:00:00     12.0    12.0
c_5     2001-07-01 00:00:00     49.0    49.0
c_6     2001-08-01 00:00:00     64.0    64.0

Internally the spatio-temporal expression will be analyzed for each time interval of the sample dataset A, the temporal functions will be replaced by numerical values, the names of the space time raster datasets will be replaced by the corresponding raster maps. The final expression will be passed to r.mapcalc, resulting in 6 runs:

r.mapcalc expression="c_1 = if(3 == 5 || 3 == 6, (a3 + b3), (a3 * b3))"
r.mapcalc expression="c_2 = if(4 == 5 || 4 == 6, (a4 + b4), (a4 * b4))"
r.mapcalc expression="c_3 = if(5 == 5 || 5 == 6, (a5 + b5), (a5 * b5))"
r.mapcalc expression="c_4 = if(6 == 5 || 6 == 6, (a6 + b6), (a6 * b6))"
r.mapcalc expression="c_5 = if(7 == 5 || 7 == 6, (a7 + b7), (a7 * b7))"
r.mapcalc expression="c_6 = if(8 == 5 || 8 == 6, (a8 + b8), (a8 * b8))"

Semantic labels present in the sample dataset A will be transferred to the output dataset.

EXAMPLES

The following command creates a new space time raster dataset january_under_0 that will set to null all cells with temperature above zero in the January maps while keeping all the rest as in the original time series. This will change the maximum values of all January maps in the new STRDS as compared to the original one, tempmean_monthly.

t.rast.mapcalc input=tempmean_monthly output=january_under_0 basename=january_under_0 \
    expression="if(start_month() == 1 && tempmean_monthly > 0, null(), tempmean_monthly)"

# print minimum and maximum only for January in the new strds
t.rast.list january_under_0 columns=name,start_time,min,max | grep 01-01
name|start_time|min|max
january_under_0_01|2009-01-01 00:00:00|-3.380823|-7e-06
january_under_0_13|2010-01-01 00:00:00|-5.266929|-0.000154
january_under_0_25|2011-01-01 00:00:00|-4.968747|-6.1e-05
january_under_0_37|2012-01-01 00:00:00|-0.534994|-0.014581

# print minimum and maximum only for January in the original strds,
# note that the maximum is different
t.rast.list tempmean_monthly columns=name,start_time,min,max | grep 01-01
2009_01_tempmean|2009-01-01 00:00:00|-3.380823|7.426054
2010_01_tempmean|2010-01-01 00:00:00|-5.266929|5.71131
2011_01_tempmean|2011-01-01 00:00:00|-4.968747|4.967295
2012_01_tempmean|2012-01-01 00:00:00|-0.534994|9.69511

Semantic label filtering

t.rast.mapcalc supports semantic label filtering similarly to t.rast.list. In example below a new STRDS will be created and filled by NDVI products.

t.rast.mapcalc inputs=test.S2_8,test.S2_4 output=ndvi basename=ndvi \
     expression="float(test.S2_8 - test.S2_4) / (test.S2_8 + test.S2_4)"

For more information about semantic label concept see i.band.library module.

AUTHOR

Sören Gebbert, Thünen Institute of Climate-Smart Agriculture

SOURCE CODE

Available at: t.rast.mapcalc source code (history)

Latest change: Thursday Jan 26 14:10:26 2023 in commit: cdd84c130cea04b204479e2efdc75c742efc4843