DESCRIPTION

The purpose of t.rast.mapcalc is to perform 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 parameter. All space time raster datasets that are referenced in the mapcalc expression must be listed in the input 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 base name 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 spatial related maps in the STRDS's are processed. Spatial 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:

Supported internal variables for relative and absolute time:

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 represent 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 represent 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.

Supported internal variables for absolute time of the current sample interval or instance:

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 variables in case of time instances.

NOTE

We will discuss the internal work of t.rast.mapcalc with an example. Imagine we have two STRDS as input, each with 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 base=c method=equal \
    expr="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 expr="c_1 = if(3 == 5 || 3 == 6, (a3 + b3), (a3 * b3))"
r.mapcalc expr="c_2 = if(4 == 5 || 4 == 6, (a4 + b4), (a4 * b4))"
r.mapcalc expr="c_3 = if(5 == 5 || 5 == 6, (a5 + b5), (a5 * b5))"
r.mapcalc expr="c_4 = if(6 == 5 || 6 == 6, (a6 + b6), (a6 * b6))"
r.mapcalc expr="c_5 = if(7 == 5 || 7 == 6, (a7 + b7), (a7 * b7))"
r.mapcalc expr="c_6 = if(8 == 5 || 8 == 6, (a8 + b8), (a8 * b8))"

EXAMPLE

Here the full script to reproduce the example:

g.region s=0 n=80 w=0 e=120 b=0 t=50 res=10 res3=10 -p3
# Lets create the raster maps
r.mapcalc  expr="a3 = 3"
r.mapcalc  expr="a4 = 4"
r.mapcalc  expr="a5 = 5"
r.mapcalc  expr="a6 = 6"
r.mapcalc  expr="a7 = 7"
r.mapcalc  expr="a8 = 8"

r.mapcalc  expr="b1 = 1"
r.mapcalc  expr="b2 = 2"
r.mapcalc  expr="b3 = 3"
r.mapcalc  expr="b4 = 4"
r.mapcalc  expr="b5 = 5"
r.mapcalc  expr="b6 = 6"
r.mapcalc  expr="b7 = 7"
r.mapcalc  expr="b8 = 8"
r.mapcalc  expr="b9 = 8"
r.mapcalc  expr="b10 = 10"
r.mapcalc  expr="b11 = 11"
r.mapcalc  expr="b12 = 12"

t.create  type=strds temporaltype=absolute \
    output=A title="Dataset A" descr="Dataset A"
    
t.create  type=strds temporaltype=absolute \
    output=B title="Dataset B" descr="Dataset B"
    
t.register  -i type=rast input=A maps=a3,a4,a5,a6,a7,a8 \
    start="2001-03-01" increment="1 months"
    
t.register  -i type=rast input=B maps=b1,b2,b3,b4,b5,b6,b7,b8,b9,b10,b11,b12 \
    start="2001-01-01" increment="1 months"

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

t.info type=strds input=C

t.rast.list -h input=C columns=name,start_time,min,max

AUTHOR

Sören Gebbert

Last changed: $Date$

DESCRIPTION

NOTE

EXAMPLE

SEE ALSO

AUTHOR