{
"cells": [
{
"cell_type": "markdown",
"id": "113058f9-8324-4368-99ed-b088987c683a",
"metadata": {},
"source": [
"# Analysis of the Vostok data\n",
"The seasonal-diurnal diagram, its hourly cross-sections, possible role of adjustments for various factors\n",
"\n",
"The source code of Figures 1.2, 1.3, 1.5, 1.S1, 1.S2 and 1.S3"
]
},
{
"cell_type": "markdown",
"id": "5870ae4e-8a7d-4d76-a7e0-a542ee0fbd25",
"metadata": {},
"source": [
"### Importing libraries"
]
},
{
"cell_type": "code",
"execution_count": 1,
"id": "c4d7641a-fe60-4177-95d8-9f10e7622f87",
"metadata": {},
"outputs": [],
"source": [
"import matplotlib.pyplot as plt\n",
"import matplotlib.transforms as tf\n",
"from matplotlib import cm, colors, colormaps\n",
"\n",
"import numpy as np\n",
"import pandas as pd\n",
"import scipy.stats as st"
]
},
{
"cell_type": "markdown",
"id": "eb2d634e-ccc5-4fd2-a9e8-05d314f12c1b",
"metadata": {},
"source": [
"### Loading source PG datasets"
]
},
{
"cell_type": "code",
"execution_count": 2,
"id": "80dbd8bc-6286-4b35-8de7-900551e1026e",
"metadata": {},
"outputs": [],
"source": [
"# potential gradient values measured at Vostok station in 1998–2001\n",
"# adjusted for local meteorological and solar wind influences\n",
"# units are V·m^(−1)\n",
"# source: Burns et al. (2012), Table 2\n",
"\n",
"vostok_old_data = (\n",
" np.array([195, 201, 205, 192, 188, 195,\n",
" 209, 198, 209, 195, 193, 192])\n",
")"
]
},
{
"cell_type": "code",
"execution_count": 3,
"id": "27eaf008-0cf7-45f6-a5ee-2b32c4c6d322",
"metadata": {},
"outputs": [],
"source": [
"# potential gradient values measured at Vostok station in 1998–2001\n",
"# units are V·m^(−1)\n",
"# source: Burns et al. (2012), Table 2\n",
"\n",
"vostok_old_data_unadjusted = (\n",
" np.array([195 - (23 - 0.1 - 1.4 + 10),\n",
" 201 - (20 - 0.1 - 1.5 + 7),\n",
" 205 - (8 - 1.5 - 0.1 + 6),\n",
" 192 - (1.0 - 1.6 + 1.5 + 4.9),\n",
" 188 - (-0.9 - 1.0 + 0.9 + 2.2),\n",
" 195 - (2.4 + 1.2 - 1.6 + 2.7),\n",
" 209 - (0.4 + 2.4 + 0.2 + 2.2),\n",
" 198 - (0.1 - 1.2 + 1.5 + 1.9),\n",
" 209 - (-0.1 - 0.1 + 1.0 + 4.9),\n",
" 195 - (8 - 0.2 + 1.7 + 4.4),\n",
" 193 - (20 + 0.1 - 0.7 + 7),\n",
" 192 - (23 + 0.2 - 0.6 + 8)])\n",
")"
]
},
{
"cell_type": "code",
"execution_count": 4,
"id": "c63b60de-82fa-4dcd-af67-a3d4738a0431",
"metadata": {},
"outputs": [],
"source": [
"# loading hourly data derived from 10-s averages\n",
"# \"Field\" column contains PG values (in V/m) without scaling and calibration\n",
"df_10s = pd.read_csv(\n",
" \"./data/Vostok/vostok_hourly_from_10_s_without_calibration_and_empty.tsv\",\n",
" sep=\"\\t\", parse_dates=[\"Datetime\"]\n",
").set_index(\"Datetime\")\n",
"\n",
"# adding a new column \"Mark\" to label this dataframe’s data\n",
"# as originating from 10-s averages\n",
"df_10s[\"Mark\"] = \"10s\""
]
},
{
"cell_type": "code",
"execution_count": 5,
"id": "d2d83b75-97ce-42ba-8a99-8597f84d7946",
"metadata": {},
"outputs": [],
"source": [
"# loading hourly data derived from 5-min averages\n",
"# \"Field\" column contains PG values (in V/m) without scaling and calibration\n",
"df_5min = pd.read_csv(\n",
" \"./data/Vostok/vostok_hourly_from_5_min_without_calibration_and_empty.tsv\",\n",
" sep=\"\\t\", parse_dates=[\"Datetime\"]\n",
").set_index(\"Datetime\")\n",
"\n",
"# Add a new column \"Mark\" to label this dataframe’s data\n",
"# as originating from 5-min averages\n",
"df_5min[\"Mark\"] = \"5min\""
]
},
{
"cell_type": "code",
"execution_count": 6,
"id": "26da1848-ad4f-4e58-8a75-d02428fb12c7",
"metadata": {},
"outputs": [],
"source": [
"# loading the earlier data set (1998-2004)\n",
"earlier_df_src = pd.read_csv(\n",
" \"./data/Vostok/vostok_1998_2004_hourly_80percent_all.tsv\",\n",
" sep=\"\\t\", parse_dates=[\"Datetime\"]\n",
").set_index(\"Datetime\")"
]
},
{
"cell_type": "code",
"execution_count": 7,
"id": "d126b8b1-0b66-4817-a556-8f33860608ae",
"metadata": {},
"outputs": [],
"source": [
"# note that all the aforementioned files assume a 80%\n",
"# data availability per hour, which means that the hourly value\n",
"# was calculated only when at least 80% of the records were \n",
"# available (at least 288 10-s averages or at least 9 5-min averages)"
]
},
{
"cell_type": "code",
"execution_count": 8,
"id": "7de3455e-ba80-420d-8fec-8f82ee28804a",
"metadata": {},
"outputs": [],
"source": [
"# combining the two dataframes:\n",
"# filling in gaps in the 10-s data with the 5-min data\n",
"df_src = df_10s.combine_first(df_5min)"
]
},
{
"cell_type": "markdown",
"id": "1f220bf9-7828-4a87-b84b-ab0b62cddcd1",
"metadata": {},
"source": [
"### Taking form factors into account"
]
},
{
"cell_type": "code",
"execution_count": 9,
"id": "dbe063a2-5d6d-4e02-b10b-1438b3f75eeb",
"metadata": {},
"outputs": [],
"source": [
"# adding a new column \"Factor\" to introduce the form factor\n",
"# by which the source values should be divided;\n",
"# the factor gradually changes from `f_0` around 1 Jan 2006\n",
"# to `f_1` around 1 Dec 2024\n",
"f_0 = 3.\n",
"f_1 = 2.4\n",
"date_range = pd.date_range(start=\"2006-01-01\", end=\"2024-12-01\", freq=\"D\")\n",
"values = np.linspace(f_0, f_1, len(date_range))\n",
"\n",
"factor_df = pd.DataFrame({\"Factor\": values}, index=date_range)\n",
"df_src = df_src.join(factor_df.resample(\"H\").ffill())"
]
},
{
"cell_type": "code",
"execution_count": 10,
"id": "2c830a98-f5f2-4792-a8eb-4e6ee3170192",
"metadata": {},
"outputs": [],
"source": [
"# Optional: one can divide the earlier data by 1.35\n",
"# to make equal the mean values of fair-weather histograms across data sets:\n",
"# earlier_df[\"Factor\"] = 1.35"
]
},
{
"cell_type": "code",
"execution_count": 11,
"id": "0b0c6844-bcb4-4331-874f-4c1eeeb91a16",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" Factor | \n",
"
\n",
" \n",
" \n",
" \n",
" | 2020-12-31 | \n",
" 2.524273 | \n",
"
\n",
" \n",
"
\n",
"