Interactive data QC

Prerequisites

• An active Solargis Analyst account.

• Installed an Analyst desktop application.

• One or more ground measurement datasets available in the Solargis Analyst application. Interactive QC is available for ground measurement datasets only.

Interactive data QC

Interactive data quality check (QC) provides an additional manual flagging of data inconsistencies that were missed by the automated QC checks. Use it to visually inspect the dataset, compare parameters against each other or against reference data, and manually flag or override the remaining inconsistencies to polish your dataset further.

The dialog opens full screen and offers three graph tabs - Sun path, Difference heatmap and Scatterplots. Each tab places its own plots above a shared time series graph, so you always keep the time context while inspecting the data from a different angle.

Key points to remember:

• Interactive data QC checks are manual, not automatic. You have to select and flag the data yourself.

• Only parameters selected for flagging can be re-flagged, and only points that already carry a flag you selected to override.

• Take advantage of the keyboard shortcuts for quick navigation when inspecting the data.

Quick start

If you already know your way around the dialog, the whole flagging cycle comes down to five steps:

1. In the sidebar menu's Data section, select one or more parameters to flag.

2. Select the flag or flags that may be overridden.

3. Select the suspicious points in any graph using the rectangle, lasso, polygon or expression selector.

4. Enter the new flag number and click “Apply” (or press Ctrl + F).

5. Click “Save”, or “Save with QC status” to mark the dataset as manually checked.

Interactive data QC tool

You can open Interactive QC from the main menu via Quality/Interactive QC or by using the quick access icon located below the main menu. The window offers the following:

1. Interactive QC tabs: Switch between the Sun path, Difference heatmap and Scatterplots views.

2. Tab graphs: The plots belonging to the active tab. Use them to spot inconsistencies that are hard to see in a time series.

3. Shared time series graph: Always visible below the tab graphs. Shows the visible parameters, theoretical, secondary and difference curves, flag overlays and maintenance markers.

4. Sidebar menu: Contains the Data, Options, and Statistics (Scatterplots only) sections. Options always mirrors the active graph tab.

5. Flagging controls: Located at the bottom of the Data section. Use them to enter and apply the flag value.

Note: Solar geometry and clear-sky index (KTm) data are derived from the site latitude, longitude and altitude when the dialog opens. By default, the first parameter from the top of the list is selected; if nothing is selected, the time series shows an empty message.

Time series graph

The bottom graph is shared by all three tabs. It displays the time series of the selected parameters in the dataset.

• Colors: Line colors come from a fixed palette assigned from the top of the parameter list downwards. They do not depend on the parameter type, and the color palette from the application settings is not used here.

• Selected vs. comparison parameters: Parameters selected for flagging are drawn at full opacity. Parameters that are only visible, but not selected, are drawn faded so you can tell them apart.

• Units: The legend shows the parameter name together with its unit.

• Empty state: The graph shows “No parameter selected.” when no parameter is visible.

Use the toolbar controls to zoom/select values in the time series for flagging.

Flags visualization

The flag table lists every known flag value with a color code and its name in the description [value] format. Each row has two toggles:

1. Select to override: Use to enable flag override and change the flag value.

2. Visibility: Toggle to visualize the flagged points on the time series in the flag color.

3. Filters menu: Select/deselect or visualize flags in bulk.

Sun path tab

The first tab places two graphs above the shared time series. Both top graphs always show the same parameter:

1. Sun path: The parameter aggregated over the sun's position, with sun azimuth on the X axis and sun elevation on the Y axis. The X axis also contains cardinal directions for easier orientation, and the southern hemisphere uses a different azimuth axis. The graph shows “No sun path data” when there is no sun path data available, for example for meteo group parameters.

2. Scalar heatmap: The same parameter by day and time of day, with the date on the X axis and the time of day on the Y axis.

Difference heatmap tab

This tab shows a single difference heatmap above the shared time series, with the date on the X axis and the time of day on the Y axis. The color represents the Reference - Estimate difference.

Scatterplots tab

This tab shows two independent scatterplots side by side above the shared time series, which is useful for cross-parameter comparison. Each shows "Timestamps … do not match" when the pair's timestamps do not overlap.Selecting points on a scatterplot requires at least one axis to come from the measured (primary) dataset.

Flagging the data manually

Interactive QC offers manual data flagging in the time series graph. The process requires close inspection of the data in the graphs. Use the toolbar controls to help you zoom and pan, and the keyboard shortcuts to navigate the charts effectively.

1. Select the parameters you want to flag in the parameter table.

2. Select the flags to be overridden in the flag table.

3. Select the points in any graph using the data selection tools. The selected area picks up values for every parameter that is checked for flagging.

4. Enter a flag value or select the flag from the list and apply it. The number of selected data points is displayed above the field.

Note: A selection only affects the parameters selected for flagging, and only the points that already carry a flag you checked for overriding. If nothing seems to happen when you apply a flag, check both tables first.

Applying the flag

Once your points are selected, you can apply a flag in any of these ways:

• Type the flag number into the field and click “Apply” or press Ctrl + F.

• Use the Choose flag button to pick a flag from the dialog and apply it. Double-clicking a row in that dialog prefills the field.

• Double-click a flag in the flag table to apply it to the selection directly.

Key points to remember:

• New flag value: If you enter a value that is not known, you are prompted to define a new custom flag.

• Selected points label: Shows how many points are selected, with a per-flag breakdown in its tooltip.

• Undo: Reverts the most recent apply. This is a single step only and asks for confirmation.

• After applying, the graphs and the flag list refresh automatically.

Copying flags from another dataset

You can copy flags from another dataset:

1. Click the copy flags button at the bottom of the sidebar.

2. In the new dialog, select the source/target dataset and parameter.

3. Select the flags to import/overwrite and click the “Import” button.

If required, you can restrict the time range of imported flags using the “Import between” row at the bottom of the window and set a nearest-timestamp tolerance between 0 and 15 minutes.

Note: When the import finishes, the number of changed rows is reported and any new flags are added to the flag table.

Using a reference dataset and theoretical curves for data comparison

Besides your own measured parameters, which are the only ones that can be flagged, three further data sources are available for visual comparison. All three can be used in the time series, the difference heatmap and the scatterplots.

Theoretical (clear-sky) curves

You can compare your solar parameters with theoretical reference curves created specifically for this comparison. They let you compare a measured parameter against an ideal sunny day based on the sun position at your site.

1. Select the theoretical parameter you want to see in the time series graph.

2. It will show as a dashed line.

One curve is available for GHI, one for DNI, and multiple GTI curves depending on the configuration of each GTI parameter. All of them are calculated together the first time any one of them is displayed.

Note: Theoretical curves are not the same as model data. Model data is satellite-based and accounts for clouds and snow, while a theoretical curve represents clear sky conditions at the given location only.

Secondary dataset parameters

If you have another dataset you want to use for the comparison - typically a model dataset - you can add it to the plots.

Any dataset in the project except the primary one can be chosen, and you select which of its parameters to load; all of them are loaded by default. The added parameters appear in the parameter list with a “(secondary)” suffix. You can remove the secondary dataset at any time using the removal button next to its name.

1. Click the button next to the secondary dataset and select the dataset in the popup.

2. Select any available secondary parameter in the parameter list to show it in the time series graph. It will be show as a dashed line.

Difference curve

You can build a curve from two parameters as parameter 1 − parameter 2, with an optional whole or decimal multiplier. The resulting equation is shown as you build it, and the pre-selection follows the parameters chosen on the Difference heatmap tab.

1. Select the “Add difference curve” from the parameter selection options menu.

2. Select reference and estimate columns, and define multiplier if required.

3. Edit the default difference curve name if required and click “Apply” to generate.

The difference curve will be added to the time series graph. You can toggle its visibility on/off as you would for any other parameter in the parameter list.

Note: A difference curve is temporary. It is not saved to the dataset and will be gone the next time you open the dialog.

Additional options and customization

Showing and filtering flag values

Two options control how flags appear in the graphs:

• Show flag values for selected parameters only: When this is on, flag values are visualized in the time series for the selected parameters only. When it is off (the default), flag values for all parameters are visible.

• Enable flag filtering: When this is on, only the checked flag values are shown across all graphs and the rest appear as gaps. You can select valid, select all or deselect all. Filtering applies to the measured (primary) parameters only - calculated and secondary curves are unaffected. With the filter off, or with all or none of the flags checked, everything is shown.

Graph interactions

Right-click any graph to open its context menu:

• Maximize / restore graph: Expands a single graph to full screen and restores it again.

• Show / hide maintenance log: Toggles the maintenance markers on the time series. The option is hidden when the dataset has no maintenance log in the displayed range.

• Add horizontal line / Add vertical line / Clear lines: Adds static reference lines to the time series graphs. Note that a line cannot be modified once created - clear the lines and add them again instead.

• Copy image / Copy all images: Copies the current graph, or all graphs, to the clipboard.

Finishing interactive data QC check

Three buttons close out your work in the dialog:

• “Save”: Saves the flag column changes together with a change log entry. It also keeps your last Sun path time series view as a bookmark, so you can continue where you left off. Any new flag values are added to the flag scheme and a success message confirms the save.

• “Save with QC status”: Does the same as Save and additionally adds the Interactive QC status to the dataset, marking it as manually checked. This removes the bookmark, so the dialog reopens with the default view next time.

• “Close”: Closes the dialog. If you have unsaved changes, you are asked to confirm first.