{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Project: Exploring Weather Trends\n",
    "\n",
    "--by Lu Tang"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Table of Contents\n",
    "<ul>\n",
    "<li><a href=\"#intro\">Introduction</a></li>\n",
    "<li><a href=\"#wrangling\">Data Wrangling</a></li>\n",
    "<li><a href=\"#eda\">Exploratory Data Analysis</a></li>\n",
    "<li><a href=\"#conclusions\">Conclusions</a></li>\n",
    "</ul>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id='intro'></a>\n",
    "## Introduction\n",
    ">The purpose of this project is to analyze data for the differences and similarities in temperature trends between global and the city Xi’an, China, which is my hometown. The project will look answers for the following questions:\n",
    ">- Is Xi’an hotter or cooler on average compared to the global?\n",
    ">- Has the difference been consistent over time?\n",
    ">- What are the overall trends for Xi’an and the world? Are they getting hotter or cooler over time?\n",
    ">- Are the overall trends consistent?\n",
    "\n",
    "**The tools used in the project:** \n",
    "\n",
    "SQL and Python, Pandas, Matplotlib and Seaborn"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id='wrangling'></a>\n",
    "## Data Wrangling"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Step 1. Getting and exploring the data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The data is stored in database, in order to understand the data, I extracted all the data from three tables using SQL, I then use Pandas to analyze and understand how the three tables are related. I discovered the following:\n",
    "- The ‘year’ column is the foreign key to relate the table of 'city_data' and 'global_data';\n",
    "- The data in the database is not sorted, but it appears like sorted from first and last few rows. *In the table of 'city_data', the 'year' starts from 1849 and it looks like ascending by year and ends at the year of 2013*. *While in the table of 'global_data', the 'year' starts from the 1750 to 2015, also ascending by year*. However after I join the two tables, I found the new table has fewer rows than the original 'city_data'. If the 'city_data' is from 1849 to 2013 as the first and last few rows show, then the joined table should have the same rows as the 'city_data', but it is not, which means the 'city_data' has longer year expand than the ‘year’ in the global_data. After sorting city_data, *I found the ‘year’ in 'city_data' starts from 1743 to 2013*. After joining two tables, the year range is from 1750 to 2013.\n",
    "- After understanding the data, I then wrote a SQL to get the data we need for this project."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**The SQL query to extra data**\n",
    "\n",
    "- SELECT c.*, g.avg_temp avg_temp_global\n",
    "- FROM city_data c\n",
    "- JOIN global_data g\n",
    "- ON c.year=g.year\n",
    "- WHERE c.city='Xian'\n",
    "- ORDER BY year \n",
    "\n",
    "Saved as 'results.csv'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Data Description:**\n",
    "- year (from 1820 to 2013)\n",
    "- city (we have selected for the city of 'xi'an')\n",
    "- country (the country is China)\n",
    "- avg_temp (the average temperature for Xian )  \n",
    "- avg_temp_y (the average temperature for global)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Step 2. Data cleaning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>year</th>\n",
       "      <th>city</th>\n",
       "      <th>country</th>\n",
       "      <th>avg_temp</th>\n",
       "      <th>avg_temp_global</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1820</td>\n",
       "      <td>Xian</td>\n",
       "      <td>China</td>\n",
       "      <td>9.55</td>\n",
       "      <td>7.62</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1821</td>\n",
       "      <td>Xian</td>\n",
       "      <td>China</td>\n",
       "      <td>11.12</td>\n",
       "      <td>8.09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1822</td>\n",
       "      <td>Xian</td>\n",
       "      <td>China</td>\n",
       "      <td>11.16</td>\n",
       "      <td>8.19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1823</td>\n",
       "      <td>Xian</td>\n",
       "      <td>China</td>\n",
       "      <td>11.76</td>\n",
       "      <td>7.72</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1824</td>\n",
       "      <td>Xian</td>\n",
       "      <td>China</td>\n",
       "      <td>NaN</td>\n",
       "      <td>8.55</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   year  city country  avg_temp  avg_temp_global\n",
       "0  1820  Xian   China      9.55             7.62\n",
       "1  1821  Xian   China     11.12             8.09\n",
       "2  1822  Xian   China     11.16             8.19\n",
       "3  1823  Xian   China     11.76             7.72\n",
       "4  1824  Xian   China       NaN             8.55"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# loading data in pandas and display first 5 rows\n",
    "import pandas as pd\n",
    "df=pd.read_csv('results.csv')\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>year</th>\n",
       "      <th>city</th>\n",
       "      <th>country</th>\n",
       "      <th>avg_temp</th>\n",
       "      <th>avg_temp_global</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>189</th>\n",
       "      <td>2009</td>\n",
       "      <td>Xian</td>\n",
       "      <td>China</td>\n",
       "      <td>12.53</td>\n",
       "      <td>9.51</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>190</th>\n",
       "      <td>2010</td>\n",
       "      <td>Xian</td>\n",
       "      <td>China</td>\n",
       "      <td>12.59</td>\n",
       "      <td>9.70</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>191</th>\n",
       "      <td>2011</td>\n",
       "      <td>Xian</td>\n",
       "      <td>China</td>\n",
       "      <td>12.08</td>\n",
       "      <td>9.52</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>192</th>\n",
       "      <td>2012</td>\n",
       "      <td>Xian</td>\n",
       "      <td>China</td>\n",
       "      <td>11.90</td>\n",
       "      <td>9.51</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>193</th>\n",
       "      <td>2013</td>\n",
       "      <td>Xian</td>\n",
       "      <td>China</td>\n",
       "      <td>14.46</td>\n",
       "      <td>9.61</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     year  city country  avg_temp  avg_temp_global\n",
       "189  2009  Xian   China     12.53             9.51\n",
       "190  2010  Xian   China     12.59             9.70\n",
       "191  2011  Xian   China     12.08             9.52\n",
       "192  2012  Xian   China     11.90             9.51\n",
       "193  2013  Xian   China     14.46             9.61"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# display last 5 rows\n",
    "df.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>avg_temp_xian</th>\n",
       "      <th>avg_temp_global</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>year</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1820</th>\n",
       "      <td>9.55</td>\n",
       "      <td>7.62</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1821</th>\n",
       "      <td>11.12</td>\n",
       "      <td>8.09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1822</th>\n",
       "      <td>11.16</td>\n",
       "      <td>8.19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1823</th>\n",
       "      <td>11.76</td>\n",
       "      <td>7.72</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1824</th>\n",
       "      <td>NaN</td>\n",
       "      <td>8.55</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      avg_temp_xian  avg_temp_global\n",
       "year                                \n",
       "1820           9.55             7.62\n",
       "1821          11.12             8.09\n",
       "1822          11.16             8.19\n",
       "1823          11.76             7.72\n",
       "1824            NaN             8.55"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# rename 'avg_temp'\n",
    "df.rename({'avg_temp':'avg_temp_xian'},axis=1,inplace=True)\n",
    "# set column 'year' as index\n",
    "df.index=df['year']\n",
    "# delete unneccesary rows\n",
    "df.drop(['year','city','country'],axis=1, inplace=True)\n",
    "#check the result\n",
    "df.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 194 entries, 1820 to 2013\n",
      "Data columns (total 2 columns):\n",
      "avg_temp_xian      179 non-null float64\n",
      "avg_temp_global    194 non-null float64\n",
      "dtypes: float64(2)\n",
      "memory usage: 4.5 KB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "avg_temp_xian      15\n",
       "avg_temp_global     0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Check sum of NaN data\n",
    "df.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Drop null since there is less then 10% of the data\n",
    "df.dropna(inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id='eda'></a>\n",
    "## Exploratory Data Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>avg_temp_xian</th>\n",
       "      <th>avg_temp_global</th>\n",
       "      <th>moving_avg_xian</th>\n",
       "      <th>moving_avg_global</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>year</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1820</th>\n",
       "      <td>9.55</td>\n",
       "      <td>7.62</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1821</th>\n",
       "      <td>11.12</td>\n",
       "      <td>8.09</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1822</th>\n",
       "      <td>11.16</td>\n",
       "      <td>8.19</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1823</th>\n",
       "      <td>11.76</td>\n",
       "      <td>7.72</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1837</th>\n",
       "      <td>21.19</td>\n",
       "      <td>7.38</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1840</th>\n",
       "      <td>10.81</td>\n",
       "      <td>7.80</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1841</th>\n",
       "      <td>10.26</td>\n",
       "      <td>7.69</td>\n",
       "      <td>12.264286</td>\n",
       "      <td>7.784286</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1842</th>\n",
       "      <td>11.05</td>\n",
       "      <td>8.02</td>\n",
       "      <td>12.478571</td>\n",
       "      <td>7.841429</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1843</th>\n",
       "      <td>11.12</td>\n",
       "      <td>8.17</td>\n",
       "      <td>12.478571</td>\n",
       "      <td>7.852857</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1844</th>\n",
       "      <td>11.01</td>\n",
       "      <td>7.65</td>\n",
       "      <td>12.457143</td>\n",
       "      <td>7.775714</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      avg_temp_xian  avg_temp_global  moving_avg_xian  moving_avg_global\n",
       "year                                                                    \n",
       "1820           9.55             7.62              NaN                NaN\n",
       "1821          11.12             8.09              NaN                NaN\n",
       "1822          11.16             8.19              NaN                NaN\n",
       "1823          11.76             7.72              NaN                NaN\n",
       "1837          21.19             7.38              NaN                NaN\n",
       "1840          10.81             7.80              NaN                NaN\n",
       "1841          10.26             7.69        12.264286           7.784286\n",
       "1842          11.05             8.02        12.478571           7.841429\n",
       "1843          11.12             8.17        12.478571           7.852857\n",
       "1844          11.01             7.65        12.457143           7.775714"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Calculating 7-year moving average for xian and global data and make it new columns\n",
    "df['moving_avg_xian']=df['avg_temp_xian'].rolling(window=7).mean()\n",
    "df['moving_avg_global']=df['avg_temp_global'].rolling(window=7).mean()\n",
    "df.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "###  Data Visualization "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Weather Trends')"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "import seaborn as sns\n",
    "sns.set(style='darkgrid', context='talk', palette='Dark2')\n",
    "\n",
    "fig,ax = plt.subplots(figsize=(16,8))\n",
    "ax.plot(df['moving_avg_xian'], label='Xian Weather Trends')\n",
    "ax.plot(df['moving_avg_global'], label='Global Weather Trends')\n",
    "\n",
    "ax.legend(loc='best')\n",
    "ax.set_xlabel('Year')\n",
    "ax.set_ylabel('Temparature')\n",
    "ax.set_title('Weather Trends')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Interpretation:\n",
    "- Xian’s average temperature is higher than global as the green line for Xian is higher; thus Xian is generally hotter.\n",
    "- The difference of temperature between Xian and global has been consistent over time as the gap between the two lines are similar over most of the period.\n",
    "- The overall trend for Xian and global are both increasing; both Xian and global are getting hotter.\n",
    "- There is an exception for the trend. In the earlier few years before 1850, Xian’s temperature was extremely high and then decreased sharply while globe’s temperature is climbing mildly during the period. There maybe some extreme weather in xian during that time, but as the period is relatively short, it does not affect our conclusion for the overall trend.\n",
    "- The increase rate of the temperature is higher in recent few years for both Xian and global. We can see from the chart, roughly after 1975, both lines are increasing quicker, however xian has higher fluctuation which global temperature is going up more stably. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Conclusion\n",
    "> In this project, I analyzed the weather trend using temperature data for Xi’an and Global from the year 1820 to 2013. We can conclude that Xi'an is a hotter place and has higher fluctuation on the weather compared with global average. However, the temperature for both Xian and global are increasing over years and particularly, it is increasing at higher rate in recent years. Furthermore, based on historical trends, we can predict that the future trend will be continuely increasing at higher rate; Our world is facing climate change and protecting the environment is very important.\n",
    "\n",
    ">**Further Notes:**\n",
    ">This project is mainly focused on EDA (Exploratory Data Analysis). To precisely predict the future trends, a robust data prediction model is needed, but this is beyond the purpose of this projects. In my other projects, there will be machine learning, data modeling and predictions. "
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}