Lab 03 - EDA and Viz 1

JSC370 (Winter 2026)

Published

January 21, 2026

Learning Goals

  • Download lab 3 here
  • Read in from github and get familiar with a meteorology dataset
  • Plan out how to tackle the objective question
  • Step through the EDA “checklist” presented in the class slides
  • Make exploratory graphs and maps

Deliverables

  • Complete the code and summaries and upload your .qmd and rendered .html to Quercus

Lab Description

We will work with similar meteorological data that was presented in lecture. Recall the dataset consists of weather station readings in the contiguous US.

The objective of the lab is to find the weather station with the lowest elevation, find out where this location is (make a map), and make a time series plot of temperature at this station.

Steps

0. Load libraries

Notes:

  • import the necessary libraries
  • note setting pandas options for nicer display of the output
  • in the yaml above set enabled to true when you want the code to render
import pandas as pd
pd.set_option("display.float_format", "{:.2f}".format) 
import numpy as np
import matplotlib.pyplot as plt
import geopandas as gpd
import contextily as cx
from IPython.display import display

1. Read in the data

Connect to github and read in data with pandas.

url = "https://raw.githubusercontent.com/JSC370/JSC370-2026/main/data/met_all_2025.gz"
met = pd.read_csv(...)

2. Check the memory size, dataset dimensions, and the header. How many columns, rows are there?

  • fill in where there are …
met.memory_usage(deep=True).sum() / ...
met.shape[...]
met.head

Summary: - add bullets with your summary

3. Look at data types of the variables in the dataset.

met.info

Summary: - add bullets with your summary

4. Take a closer look at the key variables (temp and elev).

  • Are there missing data in these variables? If so, make sure they are coded correctly.
key_vars = ['temp', 'elev']

met[...].isnull().sum()
  • Are there any unusual values that look suspicious?
minT = met['temp']
maxT = ...

minE = met['elev']
maxE = ...

pring(minT, maxT, minE, maxE)

Summary: - Add your summary here

5. Check the data against an external data source and make any other adjustments.

  • Check that the range of elevations make sense. Google or ChatGPT is your friend here.

  • Fix any problems that arise in your checks.

met.loc[...] = np.nan
met[['temp']].describe().round(4)

Summary:

cold_temps = met[met['temp'] ...]

# display data here extract lat and lon and check google

Summary:

# Fix up data here, remove implausible values

6. Answer research questions

Remember to keep the initial questions in mind. We want to pick out the weather station with minimum elevation and examine its temperature.

Some ideas for steps: 1. subset the data for the weather station with minimum elevation 2. look at histograms of temperature 3. make a time series plot (need to create date variable) 4. make a map to see where it is

6a. Subset minimum elevation site

# Subset data for the weather station with minimum elevation

6b. Histogram of temperature at minimum elevation site

plt.hist(...)

6c. Create date variable and time series plot

  • Look at the time series of temperature at this location. For this we will need to create a date-time variable for the x-axis.

  • Summarize any trends that you see in these time series plots.

# Create a date variable for min_elev_station
min_elev_station = min_elev_station.copy()
min_elev_station['date'] = pd.to_datetime(
    min_elev_station[...]
)

# Sort by date
min_elev_station = min_elev_station.sort_values('date')

# Create line plot of temperature by date

plt.plot(...)
plt.xlabel()
plt.ylabel()
plt.show()

Summarize time series plot:

6d. Where is the lowest weather station?

station_location = min_elev_station[...].drop_duplicates()

lat = station_location['lat'].iloc[0]
lon = station_location['lon'].iloc[0]
elev = station_location['elev'].iloc[0]

# make elev into GeoDataFrame for mapping
gdf = gpd.GeoDataFrame(
    {'elevation': [elev]},
    geometry=gpd.points_from_xy([lon], [lat]),
    crs='EPSG:4326'
)

# Convert to Web Mercator for basemap
gdf = gdf.to_crs('EPSG:3857')

# map
fig, ax = plt.subplots(figsize=(10, 8))
gdf.plot(ax=ax, color='red', markersize=300, marker='*', edgecolor='black', linewidth=2, zorder=5)
cx.add_basemap(ax, source=cx.providers.OpenStreetMap.Mapnik)
ax.set_xlabel('Longitude')
ax.set_ylabel('Latitude')
ax.set_title(f'Lowest Elevation Station (Elevation: {elev} m)')
plt.show()

Summarize where the lowest elevation station is located.