def get_primary_energy_consumption() -> Tuple[pd.DataFrame, List[dict]]:
print("retrieving primary energy consumption data...")
variables = [143360, 143363, 143364]
dataframes = []
codebook = []
for var_id in variables:
df, meta = get_owid_variable(var_id, to_frame=True)
assert re.search(
r"energy", meta["name"],
re.I), "'energy' does not appear in energy consumption variable"
df = df.rename(columns={
"entity": "Country",
"year": "Year",
"value": meta["name"],
}).drop(columns=["variable"])
dataframes.append(df)
codebook.append({
"name": meta["name"],
"description": meta["description"],
"source": meta["source"]["name"],
})
df = reduce(
lambda left, right: pd.merge(
left, right, on=["Country", "Year"], how="outer", validate="1:1"),
dataframes,
)
return df, codebook
def get_n2o_emissions() -> Tuple[pd.DataFrame, List[dict]]:
print("retrieving N2O emissions data...")
variables = [142812, 142848]
dataframes = []
codebook = []
for var_id in variables:
df, meta = get_owid_variable(var_id, to_frame=True)
# appends ' (n2o)' to variable name if it does not exist, because
# variable name for 142848 ("Total including LUCF (per capita)") does
# not already contain N2O (for disambiguation from other gases).
if not re.search(r"n2o", meta["name"], re.I):
meta["name"] += " (N2O)"
assert re.search(
r"n2o", meta["name"],
re.I), "'n2o' does not appear in n2o emissions variable"
# fix: if conversionFactor==1e6, then the variable is actually stored in
# million tonnes. Updates the description accordingly
description = meta["description"]
if (meta["display"]["unit"] == "tonnes CO₂e"
and meta["display"]["conversionFactor"] == 1e6):
if not re.search("million tonnes", description):
new_description = re.sub("tonnes", "million tonnes",
description)
assert (
new_description != description
and "million tonnes" in new_description
), 'Expected "million tonnes" to be present in modified description.'
description = new_description
df = df.rename(columns={
"entity": "Country",
"year": "Year",
"value": meta["name"],
}).drop(columns=["variable"])
dataframes.append(df)
codebook.append({
"name": meta["name"],
"description": description,
"source": meta["source"]["name"],
})
df = reduce(
lambda left, right: pd.merge(
left, right, on=["Country", "Year"], how="outer", validate="1:1"),
dataframes,
)
# fix: replaces "European Union (27)" with "EU-27" for consistency with CO2
# emissions dataset
df["Country"].replace("European Union (27)", "EU-27", inplace=True)
return df, codebook
def get_gdp() -> Tuple[pd.DataFrame, List[dict]]:
print("retrieving gdp data...")
df, meta = get_owid_variable(146201, to_frame=True)
df = df.rename(columns={
"entity": "Country",
"year": "Year",
"value": meta["name"]
}).drop(columns=["variable"])
codebook = [{
"name": meta["name"],
"description": meta["description"],
"source": meta["source"]["name"],
}]
return df, codebook
def get_ch4_emissions() -> Tuple[pd.DataFrame, List[dict]]:
print("retrieving CH4 emissions data...")
variables = [142803, 142841]
dataframes = []
codebook = []
for var_id in variables:
df, meta = get_owid_variable(var_id, to_frame=True)
assert re.search(
r"ch4", meta["name"],
re.I), "'ch4' does not appear in ch4 emissions variable"
# fix: if conversionFactor==1e6, then the variable is actually stored in
# million tonnes. Updates the description accordingly
description = meta["description"]
if (meta["display"]["unit"] == "tonnes CO₂e"
and meta["display"]["conversionFactor"] == 1e6):
if not re.search("million tonnes", description):
new_description = re.sub("tonnes", "million tonnes",
description)
assert (
new_description != description
and "million tonnes" in new_description
), 'Expected "million tonnes" to be present in modified description.'
description = new_description
df = df.rename(columns={
"entity": "Country",
"year": "Year",
"value": meta["name"],
}).drop(columns=["variable"])
dataframes.append(df)
codebook.append({
"name": meta["name"],
"description": description,
"source": meta["source"]["name"],
})
df = reduce(
lambda left, right: pd.merge(
left, right, on=["Country", "Year"], how="outer", validate="1:1"),
dataframes,
)
# fix: replaces "European Union (27)" with "EU-27" for consistency with CO2
# emissions dataset
df["Country"].replace("European Union (27)", "EU-27", inplace=True)
return df, codebook
def get_co2_emissions() -> Tuple[pd.DataFrame, List[dict]]:
print("retrieving CO2 emissions data...")
variables = [
179841,
179842,
179845,
179847,
179848,
179851,
179852,
179855,
179856,
179859,
179860,
179863,
179864,
179867,
179868,
179871,
179872,
179873,
179874,
179875,
179876,
179877,
179878,
179880,
179882,
179884,
179886,
179888,
179890,
179892,
179893,
179894,
179895,
179896,
179897,
179898,
179899,
179900,
179901,
179902,
179903,
179904,
179905,
179906,
]
dataframes = []
codebook = []
for var_id in tqdm(variables):
df, meta = get_owid_variable(var_id, to_frame=True)
assert re.search(
r"co2", meta["name"],
re.I), "'co2' does not appear in co2 emissions variable"
# converts tonnes to million tonnes for CO2 emissions variables with
# unit=="tonnes".
description = meta["description"]
assert meta["unit"] in [
"tonnes",
"tonnes per capita",
"%",
"kilograms per $PPP",
"kilograms per kilowatt-hour",
], (f"Encountered an unexpected unit value for variable {var_id}: "
f'"{meta["unit"]}". get_co2_emissions() may not work as '
"expected.")
if meta["unit"] == "tonnes":
assert "conversionFactor" not in meta["display"], (
f"variable {var_id} has a non-null conversion factor "
f"({meta['display']['conversionFactor']}). Variable may not "
"actually be stored in tonnes.")
df["value"] /= 1e6 # convert tonnes to million tonnes
new_description = re.sub("tonnes", "million tonnes", description)
assert (
new_description != description
and "million tonnes" in new_description
), 'Expected "million tonnes" to be present in modified description.'
description = new_description
df = df.rename(columns={
"entity": "Country",
"year": "Year",
"value": meta["name"],
}).drop(columns=["variable"])
dataframes.append(df)
codebook.append({
"name": meta["name"],
"description": description,
"source": meta["source"]["name"],
})
df = reduce(
lambda left, right: pd.merge(
left, right, on=["Country", "Year"], how="outer", validate="1:1"),
dataframes,
)
return df, codebook