Unit 26 - MODIS ST scripting

Let’s start with a simple model for computing LST statistics in Germany area for given period. The raster map is aggregated from input space-time raster dataset modis_cm by t.rast.series. Statistics is computed by r.univar.

t.rast.series input=modis_c output=t_rast_series_out \
 where="start_time >= '2023-03-01' and start_time < '2023-04-01'"
r.univar -g --overwrite map=t_rast_series_out
../_images/modis-date-stats.png

Fig. 120 Sample model to download: modis-date-stats.gxm.

In the next step the exported model into Python will be improved:

  1. The option where will be split into two options (18-29,34,60-61): start - Start date and end - End date

  2. New function check_date() to validate input dates will be added (35,37,44-47,50-52)

  3. Output of r.univar will be parsed and statistics printed (37,72,74-77)

 1#!/usr/bin/env python3
 2#
 3##############################################################################
 4#
 5# MODULE:       modis-date-stats
 6#
 7# AUTHOR(S):    martin
 8#
 9# PURPOSE:      Script generated by wxGUI Graphical Modeler.
10#
11# DATE:         Fri Feb  2 18:04:47 2024
12#
13##############################################################################
14
15# %module
16# % description: Computes LST stats for given period (limited to Germany and 2023).
17# %end
18# %option 
19# % key: start
20# % description: Start date (eg. 2023-03-01)
21# % type: string
22# % required: yes
23# %end
24# %option 
25# % key: end
26# % description: End date (eg. 2023-04-01)
27# % type: string
28# % required: yes
29# %end
30
31import sys
32import os
33import atexit
34from subprocess import PIPE
35from datetime import datetime
36
37from grass.script import parser, fatal, parse_key_val
38from grass.pygrass.modules import Module
39
40def cleanup():
41    Module("g.remove", flags="f", type="raster",
42           name="t_rast_series_out")
43
44def check_date(date_str):
45    d = datetime.strptime(date_str, '%Y-%m-%d')
46    if d.year != 2023:
47        fatal("Only year 2023 allowed")
48
49def main(options, flags):
50    check_date(options['start'])
51    check_date(options['end'])
52
53    Module("t.rast.series",
54           overwrite=True,
55           input="modis_c",
56           method="average",
57           order="start_time",
58           nprocs=1,
59           memory=300,
60           where="start_time >= '{}' and start_time < '{}'".format(
61                 options["start"], options["end"]),
62           output="t_rast_series_out",
63           file_limit=1000)
64
65    m = Module("r.univar",
66               flags='g',
67               overwrite=True,
68               map="t_rast_series_out",
69               percentile=90,
70               nprocs=1,
71               separator="pipe",
72               stdout_=PIPE)
73
74    stats = parse_key_val(m.outputs.stdout, val_type=float)
75    print('Min: {0:.1f}'.format(stats['min']))
76    print('Max: {0:.1f}'.format(stats['max']))
77    print('Mean: {0:.1f}'.format(stats['mean']))
78
79    return 0
80
81if __name__ == "__main__":
82    options, flags = parser()
83    atexit.register(cleanup)
84    os.environ["GRASS_OVERWRITE"] = "1"
85    sys.exit(main(options, flags))

Sample script to download: modis-date-stats.py

Example of usage:

modis-date-stats.py start=2023-03-01 end=2023-04-01

Min: -10.6
Max: 11.5
Mean: 3.7