Style Guidelines for Time Series Plots

[ashenfernando1]

an example time series plot:

• change date format to YYYY/MM/DD (the month plots will be YYYY/MM, and the annual plots will be YYYY)
• larger tick labels on X & Y
• Horizontal time labels
• Time ticks every hour
• Time labels every 3 hours
• Very fine grid lines in background every in X
• Intermediate unlabeled y-tick marks
• X & Y titles bigger, 1pt smaller than title font
• Add lines to close graph
• Format Lat and Long and add 2 decimal places (Lat $^{\circ}$, Long $^{\circ}$)
• marginal PDF needs an x-axis with labels, maybe in %
• Very fine grid for PDF
• plot scatter points to represent the uncertainty
• x-label title (local time zone in HH:mm)
• turn all this into a function that takes dataframe, pm_size

[john-waczak]

Once we have agreed on designs for teach type of time series plot, we should develop a set of template files for Python, Julia, R, etc... with functions to generate graphs so we can reduce the need for boilerplate code. It would be nice to put these into a separate repo (perhaps MintsPlotRecipes) so we can reuse them for multiple projects

[john-waczak]

I'll start by copying over my Julia ML plotting recipes and style files.

[ashenfernando1]

if you include a list of check marks on the first post of the issue page, it'll interpret as tasks to be completed. neat! @john-waczak

also i think im enjoying just tagging people now

[davidlary]

Thanks!

Please can we use the date format YYYY/MM/DD (the month plots will be YYYY/MM, and the annual plots will be YYYY).

Please make tick labels on X & Y larger.

Please can the Time labels be horizontal
Time ticks every hour
Time labels every 3 hours
Very fine graph paper in background every hour in x direction

Intermediate unlabeled y-tick marks

For the marginal PDF we should probably have an x-axis with labels, maybe in % (any thoughts?)
Would be good to have a very fine graph paper in background for PDF

Please may X & Y axis titles bigger, just 1 pt smaller than the title font. For the y-axis you can leave out the word concentration.

Perhaps having an upper x-axis and a right y-axis on each panel to close the box may look better (any thoughts?)

In the title when you say at Lat, Long, use parenthesis ie ... at (Lat°,Long°)with the lat & long in degrees with a degree symbol ° and probably use just 2 decimal places for the lat and long.

[ashenfernando1]

thanks, will update these as tasks

[davidlary]

Icons for time series plots.

Perhaps use these icons for each hour of the day on daily plots

Rainfall

Rainfall intensity is typically classified based on the rate of precipitation in millimeters per hour (mm/hour). Here are some general classifications:

1. Light Rain: Less than 2.5 mm/hour
2. Moderate Rain: Between 2.5 mm/hour and 7.6 mm/hour
3. Heavy Rain: Greater than 7.6 mm/hour

Wind speed

Wind speed is classified into different categories in meters per second (m/s):

1. Light Wind: 0.3 to 3.3 m/s (1 to 11.9 km/h or 0.7 to 7.4 mph)
2. Moderate Wind: 3.4 to 7.9 m/s (12 to 28.4 km/h or 7.5 to 17.7 mph)
3. Strong Wind: 8.0 to 13.8 m/s (28.8 to 49.7 km/h or 17.8 to 30.8 mph)
4. Very Strong Wind: 13.9 to 20.7 m/s (50 to 74.5 km/h or 30.9 to 46.3 mph)

Cloudiness

Based on Lakitha's coud fraction we can use icons fror cloudiness, I have created icons for every 10%

Lightning

If we detect lighting we can display a lightning icon

[john-waczak]

@davidlary @Vinukcs Idea: add "context bars" above or below the plot with icons

[john-waczak]

@Vinukcs Create a custom matplotlib stylesheet to set default values for plots using matplotlib.

• my custom Makie.jl (Julia) default plot settings (for your reference): matplotlib style sheets: https://matplotlib.org/stable/users/explain/customizing.html#defining-your-own-style
• page detailing matplotlib stylesheets: matplotlib style sheets: https://matplotlib.org/stable/users/explain/customizing.html#defining-your-own-style
• font files for the Montserrat font are in /font-files
• for setting the font: https://matplotlib.org/stable/api/font_manager_api.html