Graphing weather statistics
with RRDTool

---

1.
Loading weather information for Varkaus Kosulanniemi [FMISID=101421].
10/2/2019 - The use of open data from the Finnish Meteorological Institute no longer requires registration.

wget -t 1 -O ~/fmidata/varkaus.xml ${URL}
URL="http://opendata.fmi.fi/wfs?
service=WFS^version=2.0.0^
request=getFeature^storedquery_id=fmi::observations:: weather::timevaluepair^fmisid=101421^timestep=10^"

Ampersand characters have been replaced with ^

---

2.
Parsed XML data to time:value pairs:
0 15-07-2017 13:50 Temperature 17.1 °C
1 15-07-2017 13:50 Wind 3.4 m/s
2 15-07-2017 13:50 Direction 311.0°
3 15-07-2017 13:50 Humidity 57.0 %
4 15-07-2017 13:50 Dew Point 8.6 °C
5 15-07-2017 13:50 Pressure 1013.7 hPa
6 15-07-2017 13:50 Felt like 16.9 °C

Save values to CSV file for RRDs tool input.
Save values to MySQL database, just in case.

---

3.
Collect data from CSV file:
TEMP=`cat ~/fmidata.csv | cut -d ',' -f1`
WIND=`cat ~/fmidata.csv | cut -d ',' -f2`
... and so on ..

Update RRD database:
rrdtool update ${RRDDIR}/temp.rrd N:${TEMP}
rrdtool update ${RRDDIR}/wind.rrd N:${WIND}

---

4.
Create RRD Ghraps (Daily):
rrdtool graph $OUTPATH/wind-day.png
--start -24h --end N --step 300
...
DEF:w=$RRDPATH/$WINDFILE:wind:AVERAGE
LINE2:w$WINDCOLOR:"$WINDWORD"
COMMENT:"5 $MINUTEWORD $avgWORD\\n"
GPRINT:w:MIN:"$MINWORD\: %5.1lf"
GPRINT:w:MAX:"$MAXWORD\: %5.1lf"
GPRINT:w:AVERAGE:"$AVGWORD\: %5.1lf"
GPRINT:w:LAST:"$LASTWORD\: %5.1lf\\n"

---

5.
Save FMI data to Windrose database.
There are eight "buckets" (0-7) with ranges
displayed on the legend (m/s):
0-1, 1-2, 2-3, 3-5, 5-6, 6-10, 10-13.5, 13-99

The first element (0) specify the percentage of 0-1 reading the sum of these reading,
this will be placed in the center of the windrose plot.
The second element (1) specifies the percentage of 1-2 readings,
the third element (2) the percentage of 2-3 readings and so on.

First create a new windrose graph:
$graph = new WindroseGraph(600,600);
Get data from database and create the windrose plot: $wp = new WindrosePlot($data);

---

6.
Copy FMI and Windrose graphs to server.
Recurse through a local FMI directory and upload the entire contents to a remote FTP server.
Copy the entire "FI101398-all/" directory located at ~/fmidata/ to a remote server's /public_html/fmidata/FI101398-all/ via FTP.

---

Data Collection System (24.2.2022)
  Operating system: Linux
  Kernel: 5.4.0-100-generic
  Machine: x86_64 (Ubuntu 20.04.4 LTS Server)
  Server version: Apache/2.4.41 (Ubuntu)
  Data Collection: Python 3.8.10
  Graph: RRDtool 1.7.2
  The windrose graphic: jpgraph-3.1.x