Layer Visualization | Technosylva Help Center

Layer Visualization Options

To access the Layer Visualization panel, click the three dots options menu (⋮) next to any layer and select Layer Visualization:

Layer Visualization panel dropdown selection

Not all data layers have the same options, so the dropdown may display fewer options depending on the layer selected.

The Layer Visualization panel is a comprehensive styling interface that allows users to customize the visual appearance of each data layer.

Example of the layer visualization panel

Options when using the Data Visualization panel include:

Calculate using dropdown to select the dataset (e.g. circuit or segment)
Select an attribute dropdown to select the desired attribute/metric
Styling Options (tabs):
- Color: Multiple color scheme options including gradients
- Size: Adjust feature sizes
- Opacity: Control transparency
- Defaults: Save frequently used styles as defaults

Styling By Color

Color Options

A variety of color gradients are available to choose from.

The choice of color gradient significantly impacts how effectively the data communicates risk, patterns, and insights.

Green-Yellow-Red (Traffic Light)

Best for: Risk assessment, urgency indicators, performance metrics

This is the most intuitive gradient for risk visualization because it leverages universal color associations:

Green = Safe, low risk, good
Yellow = Caution, moderate risk, warning
Red = Danger, high risk, critical

Use when: Visualizing wildfire risk, circuit overload probability, safety scores, or any metric where the intent is to invoke an immediate understanding of "good vs. bad." This is likely the default choice for utility risk composites because stakeholders can instantly identify high-risk areas.

Other Color Options

Green/Teal Gradients

Best for: Positive/neutral metrics
Use when: Any metric where there is no intent to trigger alarm.

Peach/Orange Gradients

Best for: Moderate emphasis
Use when: Showing characteristics requiring emphasis but not alarm.

Blue-Pink/Purple (Diverging)

Best for: Showing deviation from a midpoint, change analysis
Use when: Displaying changes from baseline, deviation from normal conditions, or comparative analysis where you need to show both directions of change.

Blue Gradients

Best for: Continuous scales, water resources
Use when: Showing water resources or any metric where blue is semantically appropriate.

Purple/Lavender Gradients

Best for: Alternative emphasis, specialty metrics
Use when: a distinct color scheme that won't conflict with other risk layers is required.

Gray Gradients (Monochrome)

Best for: Neutral data, base layers, background information
Use when: Displaying reference infrastructure, historical data, or context layers that support but don't compete with primary risk visualizations.

Multi-Color Gradients (Rainbow-like)

Best for: Categorical differences, maximum distinction
Use when: Displaying categorical data where classes are different types rather than different magnitudes.

Practical Tips for FireSight

Consistency is key: Use the same color scheme for the same type of metric across different layers
Consider colorblind users: Red-green combinations in particular can be problematic for some users
Layering strategy: Use complementary color schemes when overlaying multiple layers (e.g., red for risk and blue for infrastructure)
Context matters: Consider the audience of the map (for example, public-facing maps may need softer colors than internal operational dashboards)

For wildfire risk analysis, the green-yellow-red gradient remains the gold standard because it requires no explanation—everyone immediately understands that red means high risk and requires attention.

Class Break Options

The classification method selected fundamentally changes the story the data tells. Two maps of identical data can look completely different and lead to opposite conclusions depending on the classification technique. For example, a viewer of a risk map will assume:

Red areas = high risk (top category)
Green areas = low risk (bottom category)
Equal visual weight = equal importance

However, the classification method determines what qualifies as "high" or "low," which can dramatically alter decision-making.

Stats

Styling by Color defaults to the Stats classification method.

This method uses statistical measures to create class breaks based on key statistical values:

Minimum value
Mean (average)
Selected percentiles (90th, 95th, 98th)
Maximum value

This approach highlights data distribution relative to statistical benchmarks, making it easy to identify values that fall into specific statistical categories (like "above the 95th percentile"). It's particularly useful for emphasizing outliers or specific risk thresholds.

Viewer perception:

✅ Clear risk communication: "The red areas represent the top 2% high risk location"
✅ Defensible decisions: "Resources were prioritized for circuits above the 95th percentile"
⚠️ Can minimize perception of widespread problems: If 90% of data falls below the 90th percentile, those areas all look "safe" even if they still have significant risk
⚠️ Sensitive to data distribution: If most values cluster near zero with a few extreme outliers, the majority of the map will appear very similar

Best for:

Resource allocation based on predefined risk thresholds
Regulatory reporting where identifying top-risk areas is required
Simplified dashboards where stakeholders want to see "where's the worst risk?"

Jenks (Natural Breaks)

The Jenks (Natural Breaks) algorithm minimizes variance within classes while maximizing variance between classes. It identifies natural groupings in the data by:

Finding "breaks" where there are large jumps in values
Creating classes where values within each class are as similar as possible
Ensuring maximum distinction between different classes

This method is excellent for data that clusters naturally and when you want the classification to reflect the actual distribution patterns in your data.

Viewer perception:

✅ Statistically honest: Class breaks occur where there are real differences in the data
✅ Reveals natural groupings: If circuits fall into distinct risk categories, Jenks will find them
✅ Balanced approach: Neither as extreme-focused as Stats nor as artificial as Quantiles
⚠️ Less predictable: Without knowing the data distribution, viewers can't know what "red" means (top 5%? top 20%?)
⚠️ Harder to explain: "The break at 0.573 was algorithmically determined" is less intuitive than "95th percentile"

Best for:

Data exploration when predefined thresholds don't exist
Diverse datasets where letting the data "speak for itself" is preferred
Technical audiences who understand statistical classification
Situations where natural clustering exists (e.g., urban vs. rural circuits might naturally separate)

Quantiles

Quantiles divide the data into equal-sized groups based on the number of features:

Each class contains approximately the same number of features
For example, if six classes are selected, each class will have about 1/6 of all features

This method ensures balanced representation across all classes, making it useful when you want to ensure no class is over- or under-represented. However, it can sometimes group very different values together if the data distribution is skewed.

Viewer perception:

✅ Shows relative rankings: "These circuits are in the top third, middle third, bottom third"
✅ Reveals spatial patterns: Makes it easier to see geographic clustering when data is skewed
⚠️ Can be misleading: Two values that are very close (0.451 vs 0.452) might be in different color classes, while two distant values (0.1 vs 0.3) might be in the same class
⚠️ Exaggerates differences: When data is tightly clustered, small differences look dramatic
⚠️ Downplays extremes: Truly exceptional high-risk areas don't stand out if they're grouped with moderately high values

Best for:

Comparative analysis: "Which areas are relatively higher risk than others?"
Identifying patterns in highly skewed data
Initial exploration to understand spatial distribution

Jenks-Log

This applies the Jenks Natural Breaks algorithm to log-transformed data:

First, the data is transformed using a logarithmic scale
Then Jenks classification is applied to the transformed values
The breaks are then converted back to the original scale

This method is ideal for data with exponential distributions or data that spans several orders of magnitude. It helps visualize patterns in data where small values are densely packed and large values are sparse.

Viewer perception:

✅ Reveals patterns in skewed data: Makes sparse high values visible without washing out low values
✅ Good for exponential phenomena: Fire spread, cascading failures, viral propagation
⚠️ Counterintuitive: Viewers often don't understand logarithmic scales
⚠️ Can minimize real risk differences: A circuit at 0.6 and 0.8 might look similar on log scale, but that 0.2 difference could be operationally critical

Best for:

Data with exponential characteristics
Very wide-ranging values (0.001 to 1000)
Scientific audiences familiar with log scales

Quantiles-Log

This applies Quantile classification to log-transformed data:

First, the data is transformed using a logarithmic scale
Equal-sized groups are then created based on the log-transformed values
The breaks are then converted back to the original scale

This is useful for skewed distributions where equal representation across classes is important but the data has a wide range with many small values and few large values. It combines the benefits of both log transformation and equal-frequency grouping.

Viewer perception:

✅ Equal visual weight with proportional grouping: Each color appears on the same number of features, but the grouping respects exponential relationships
✅ Reveals low-end patterns: Makes distinctions visible in the densely-packed low values that regular Quantiles might blur together
✅ More predictable than Jenks-Log: The analyst knows each class will have the same count of features
⚠️ Counterintuitive break placement: The breaks don't evenly divide the number line
⚠️ Exaggerates low-end differences and compresses high-end differences: A circuit at 0.011 and one at 0.013 might be in different classes, even though the absolute difference is tiny
⚠️ Still has the misleading proximity problem: Like regular Quantiles, values just above and below a break point might be colored very differently despite being nearly identical

Practical Tips for FireSight

In general, for risk metrics, the Stats method is often preferred because:

It allows the viewer to clearly identify high-risk areas using percentile thresholds (90th, 95th, 98th percentile)
It aligns with how organizations make decisions and allocate resources
It provides clear justification for and prioritization of projects in a way that is explainable to non-technical stakeholders through data (e.g. “vegetation management and system hardening can be justified for areas in the 98th percentile of population impacted even when when permits and/or funding are hard to obtain”)
It doesn't artificially inflate or deflate risk perception

Therefore, this is the default view when styling a layer by color.

However, the Quantiles method is recommended composite indices such as the Utility Risk Vegetation Composite and the Utility Risk Composite in FireSight. Composites have already been normalized and standardized using a quantile transformation.

The analyst should always consider the audience, purpose, and data characteristics before choosing a classification method. Regardless of the choice, the best practice is to document the classification method and break values in legends or documentation.

Additional Controls

Additional functionality when styling by color

Additional options when styling the data by color include:

"Invert Colors" button inverts the current color scheme with existing breaks
Options menu (⋮) allows individual class customization

Individual Class Customization Drop-Down Menu

Note that at this time breaks do not recalculate when changed. Therefore, a user must adjust breaks manually when adding, moving, or removing a stop.

Manual editing of Values and Labels for each class.
"Show in the legend?" toggle allows the user to select whether to show the symbology in the legend
"Reset" button resets all visualization options to default style
"Save as my default style" button saves the current configuration as the default style for this layer

Styling By Size

This panel allows the user to control the size of features on the map based on data values, creating a visual representation where feature size corresponds to a specific attribute.

Options when styling the data by size include:

Calculate using dropdown to select the dataset (e.g. circuit or segment)
Select an attribute dropdown to select the desired attribute/metric
“Enable sizing” checkbox toggles whether size-based visualization is active.
“Use the selected Metric” checkbox
- When enabled, the system uses the same attribute selected in the main "Select an attribute" dropdown
- When unchecked, you can choose a different field for sizing independent of the color/classification attribute (e.g., color by risk, size by length)
Value range controls
- By default, features are continuously sized with a minimum to maximum value
- Options menu (⋮) - Additional settings for this break allows the user to add, move, or remove stops

The user can can manually adjust values, sizes, or labels for each value line
“Show in the legend?” toggle switch controls whether the size information appears in the map legend
Reset reverts all data visualization settings to default values
“Save as my default style” button saves the current configuration as the default style for this layer

Styling By Opacity

This panel allows the user to control the opacity of features on the map based on data values, creating a visual representation where feature opacity corresponds to a specific attribute.

Options when styling the data by opacity include:

Calculate using dropdown to select the dataset (e.g. circuit or segment)
Select an attribute dropdown to select the desired attribute/metric
“Enable opacity” checkbox toggles whether opacity-based visualization is active.
“Use the selected Metric” checkbox
- When enabled, the system uses the same attribute selected in the main "Select an attribute" dropdown
- When unchecked, you can choose a different field for sizing independent of the color/classification attribute (e.g., color by risk, opacity by a second metric)
Value range controls
- By default, features are continuously made opaque with a minimum to maximum value
- Options menu (⋮) - Additional settings for this break allows the user to add, move, or remove stops

The user can can manually adjust values, opacity percentage, or labels for each value line
“Show in the legend?” toggle switch controls whether the opacity information appears in the map legend
Reset reverts all data visualization settings to default values
“Save as my default style” button saves the current configuration as the default style for this layer

Defaults

The defaults menu allows the user to apply quick styling to a layer.

Example of a default panel for a line layer

The example shown is a line layer. Polygons will have additional options.

Options when using the Defaults panel include:

Calculate using dropdown to select the dataset (e.g. circuit or segment)
Select an attribute dropdown to select the desired attribute/metric
Color selector with hue and transparency options

Line style (e.g. solid, dash, dot)
Line cap (round, butt, square)
Line width (.3-10 scale)
Reset reverts all data visualization settings to default values
“Save as my default style” button saves the current configuration as the default style for this layer

Default styles apply only to the user, not to the organization. Currently, there is no option to set organizational defaults.