Figure editor

The figure editor makes diagrams of the results, the input and measured data for reports, papers and presentations: with a title, subtitle and caption, several panels, axis titles with sub- and superscripts, annotations, a choice of fonts, colours and line styles, and the export as PNG or SVG. It opens in a window of its own with Results → Figures in the list on the left or View → Figures (new window). The figures belong to the project: they are saved in the .h1dr file (the HYDRUS-1D project files cannot store them).

The window

Every change shows at once in the preview. A click on a panel of the preview selects it for the settings on the left. F1 or ? shows this topic.

Building a figure

1. New figure ▾ offers templates: water content and pressure head profiles, concentration profiles, observation nodes, breakthrough curves, surface and bottom fluxes, the cumulative water balance, the atmospheric forcing, the retention curve and conductivity, the mass balance error, simulated against measured data, observed and fitted values of a parameter estimation, the comparison of runs and an overview with four panels. Templates that need other data (solutes, observation nodes, measured data, kept runs) are greyed out.
2. Change the size, the title and the look in the section Figure.
3. Choose the data of each panel in Layers (data): the source, the x and the y variable.
4. Adjust the axes, add annotations, and export.

Figure

The size is the size in the document: the fonts are given in points (1 pt = 1/72 inch ≈ 0.35 mm) and keep their size whatever the resolution. Journals usually ask for text of at least 6–8 pt at the printed size; the defaults (8.5 pt for the numbers, 10 pt for the axis titles) fit a figure of 85–170 mm.

The series of all layers of a panel take the colours of the palette one after the other: two layers with one series each get the first and the second colour. Viridis and Grey spread their range over all series of the panel.

The palettes follow the advice to use colours that remain distinguishable for readers with a colour vision deficiency and in print Crameri et al. (2020): Okabe–Ito, the default, is the palette of Okabe and Ito (2008), recommended for scientific figures by Wong (2011). Viridis Smith and van der Walt (2015) is a perceptually uniform sequence for series in an order such as print times or depths. Grey suits black-and-white print.

One palette per panel

With several panels, Same palette for all panels (on at first) gives the whole figure the palette chosen in the section Figure. Off, every panel has its own palette, e.g. Viridis for the profiles at the print times and Okabe–Ito for the fluxes beside them: it is chosen in Layers (data) under the geometry of the layer, and applies to all layers of the selected panel. When the option is switched off, every panel starts with the palette of the figure. With one panel (1 × 1) the option is not shown and the palette of the figure applies.

Custom colours

With the palette Custom (own colours) the series take the colours of a list of your own, in order (after the last colour the list starts again). The list is shown under the palette: in the section Figure, or in Layers (data) when the panels have their own palettes (each panel then has its own list). Edit next to a palette turns it into a custom palette with the same colours, so the figure looks the same until a colour is changed: the cycle of Okabe–Ito or Tableau 10, or one colour per series from Viridis or Grey (as many as the panel with the most series has). When Custom is chosen in the list, an empty list starts with the colours of the palette chosen before.

Each row shows the legend labels of the series that take the colour in the selected panel (resting the mouse on them lists the series of all panels; not used when no series takes it), the swatch and the code; ⬆ ⬇ change the order, 🗑 removes a colour, + Add colour adds one.

A double click on a legend label renames the series in the legend of the selected panel, e.g. "Cumulative infiltration" to "Infiltration": Enter or a click elsewhere applies the name, Escape cancels it, and an empty name brings back the automatic label. The name belongs to the series (it is stored with the automatic label of the series) and is kept when the colours change; the Legend label of a layer, with {item}, {run} and {y}, names all series of the layer at once.

A click on a swatch opens the colour picker, as in the developer tools of web browsers: the field sets the saturation (left to right) and the brightness (top to bottom), the sliders below the hue and the opacity (alpha); the code can be typed as #rrggbb or #rrggbbaa, and R, G, B, A as numbers 0–255. The same picker sets the colour of an annotation.

Font and text sizes

Font offers Noto Sans, bundled with the program, and the font families installed on the computer. Noto Sans has all the symbols of soil physics: Greek letters, sub- and superscript digits and letters, the mathematical signs and arrows (from Noto Sans Math). Characters that a chosen system font lacks are taken from Noto Sans, so θ, ≤ or ⁻¹ always appear. The sizes of the title, subtitle, caption, panel titles, axis titles, tick labels and legend are set separately.

Text: markup and symbols

All texts (titles, axis titles, legend labels, annotations) understand the markup of this help:

A line break in the title, the caption or an axis title starts a new line. The section Symbols and markup has buttons for the common symbols (θ, α, ψ, Δ, °, ±, ≤, √, ∂, →, ², ⁻¹, ₛ …): a click copies the symbol, Ctrl+V pastes it.

Panels and layers

A panel is one diagram with its axes, legend and annotations. Its title (e.g. "(a) Loam"), the position of its legend (a corner inside the plot area, outside on the right, below, or none) and the Legend width are set in the section Panel; with several panels the row of numbers selects the panel. The legend width is the width of the legend in percent of the width of the plot area (for a legend outside on the right: of the plot area without it): at 90 % the legend spans almost the whole plot area. The entries are arranged in rows of as many columns as fit, and labels too long for one column are wrapped at blanks (a word longer than the column stays whole). auto (0, at first) makes the legend as wide as its entries, in one column (below the plot: in rows across the panel); labels are wrapped only where the legend would be wider than the plot area (on the right: than half of it).

A panel shows one or more layers. A layer takes its x and y values from two columns of a data source; it draws one series for each selected item (observation node, print time, material, data set) and run. The check box shows or hides a layer, ⬆ ⬇ change the drawing order (later layers are drawn on top), 🗑 removes it, + Add layer adds a copy of the selected one.

Data sources

The menus of the x and y variables list the columns with their units, grouped (time and position, water, cumulative water fluxes, solutes, heat, numerics). When the source changes, the variables stay if the new source has them. Scale factors multiply the values (10 for cm → mm, −1 to plot downward fluxes as positive, 1/24 for hours → days).

The depth increases downward: the templates of profiles reverse the y axis (Reversed in the axis settings).

Style

Axes

The section Axes has the x axis, the y axis and the right y axis of the selected panel:

Annotations

Annotations add a text, a horizontal line (e.g. field capacity, a water table), a vertical line (the start of irrigation, a print time) or an arrow with a text to the selected panel. Their position is either in the units of the axes or relative to the plot area (0–1 from the bottom left; such annotations stay in place when the data change). With 🖱 Place with the mouse a click in the panel moves the selected annotation there; a right click sets the end of an arrow. Black annotations follow the text colour of the theme.

Data files

Import table… reads a text or CSV file as a data source of the figure, e.g. measurements or the results of another model:

time (d); theta (-)
0.5; 0.231
1.0; 0.245

Columns are separated by tabs, semicolons, commas or blanks (with semicolons a decimal comma is allowed); the first line with text gives the column names, and a unit in brackets at the end of a name becomes the unit of the axis title. Lines starting with # are ignored. The table is stored with the figure in the project.

Export

An SVG file names the fonts of its texts; where Noto Sans is not installed another font is used. SVG: text as outlines writes the glyphs as shapes instead: the file then looks the same everywhere but the text cannot be edited. The PNG, the copied image and SVG files with outlines contain the shapes of the letters of the bundled fonts, which their license permits (see Credits and license).

The figures are drawn by the program itself, the same way for the preview, the PNG and the SVG: a figure consists of layers of data drawn with a geometry against scales, as in the grammar of graphics Wilkinson (2005) and ggplot2 Wickham (2016), whose themes and legend positions the editor also follows.

See also

• Comparison of runs: several runs in one figure.
• Measured data: data shown with the results and in the figures.
• Results pages: the quick plots of the results.

Literature cited in this topic

• Crameri, F., G. E. Shephard, and P. J. Heron (2020). The misuse of colour in science communication. Nature Communications 11: 5444. doi:10.1038/s41467-020-19160-7
• Heckbert, P. S. (1990). Nice numbers for graph labels. In: A. S. Glassner (ed.), Graphics Gems, Academic Press, Boston, 61–63.
• Okabe, M., and K. Ito (2008). Color Universal Design (CUD): How to make figures and presentations that are friendly to colorblind people. J*Fly, Department of Molecular Biology, University of Tokyo. https://jfly.uni-koeln.de/color/
• Smith, N. J., and S. van der Walt (2015). A Better Default Colormap for Matplotlib. SciPy 2015 conference, Austin, TX. https://bids.github.io/colormap/
• Talbot, J., S. Lin, and P. Hanrahan (2010). An extension of Wilkinson's algorithm for positioning tick labels on axes. IEEE Transactions on Visualization and Computer Graphics 16(6): 1036–1043. doi:10.1109/TVCG.2010.130
• Wickham, H. (2016). ggplot2: Elegant Graphics for Data Analysis, 2nd edition. Springer, Cham, 260 pp. doi:10.1007/978-3-319-24277-4
• Wilkinson, L. (2005). The Grammar of Graphics, 2nd edition. Springer, New York, 690 pp. doi:10.1007/0-387-28695-0
• Wong, B. (2011). Points of view: Color blindness. Nature Methods 8(6): 441. doi:10.1038/nmeth.1618

All sources: References.