Support Quality Controlled Rain Gauge Data from KNMI Data Platform

[martinvonk]

KNMI Data Platform provides Quality Controlled Rain Gauge Data.

We can read KNMI Data Platform with NLMOD and the Quality Controlled Rain Gauge Data with the FEWS reader in HydroPandas. I am not suggesting that we include this per se. But it could be a nice addition to the existing KNMI data. I could not find any metadata on the stations locations so that is a problem if we don't know the locations beforehand..

Example Script:

# %%
import nlmod
import nlmod.read.knmi_data_platform as kdp
import pandas as pd
from xml.etree import ElementTree as ET
from pathlib import Path
nlmod.util.get_color_logger("INFO")
import pastastore as pst
import matplotlib.pyplot as plt


#%%
dataset_names = [
    "watercompany_raingauge_quality_controlled_waternet",
    "watercompany_raingauge_quality_controlled_pwn",
    "waterboard_raingauge_quality_controlled_rijnland",
    "waterboard_raingauge_quality_controlled_noorderzijlvest",
    "waterboard_raingauge_quality_controlled_limburg",
    "waterboard_raingauge_quality_controlled_hhnk",
    "waterboard_raingauge_quality_controlled_delfland",
    "waterboard_raingauge_quality_controlled_dommel",
    "waterboard_raingauge_quality_controlled_hunzeenaas",
    "waterboard_raingauge_quality_controlled_aaenmaas",
    "waterboard_raingauge_quality_controlled_hdsr"
]

dataset_name_all = "waterboard_raingauge_quality_controlled_all_combined"

# %%
dataset_name = dataset_name_all
dataset_version = "1.0"
files = kdp.get_list_of_files(
    dataset_name=dataset_name,
    dataset_version=dataset_version,
)

# %%
kdp.download_files(
    dataset_name=dataset_name,
    dataset_version=dataset_version,
    fnames=files[-1:],
    dirname=dataset_name,
    api_key=own_api_key
)

#%%
dfiles = list(Path(dataset_name).glob("*.xml"))
tree = ET.parse(dfiles[0])
root = tree.getroot()

obsd = {}
for item in root:
    if item.tag.endswith("series"):
        header = {}
        date = []
        time = []
        events = []
        for subitem in item:
            if subitem.tag.endswith("header"):
                for subsubitem in subitem:
                    prop = subsubitem.tag.split("}")[-1]
                    val = subsubitem.text
                    if prop == "x" or prop == "y" or prop == "lat" or prop == "lon":
                        val = float(val)
                    header[prop] = val
            elif subitem.tag.endswith("event"):
                date.append(subitem.attrib.pop("date"))
                time.append(subitem.attrib.pop("time"))
                events.append({**subitem.attrib})
        index = pd.to_datetime(
            [d + " " + t for d, t in zip(date, time)], errors="coerce"
        )
        ts = pd.DataFrame(events, index=index)
    obsd[header["locationId"]] = {"header": header, "series": ts}