Using the Alaska Wildfires Web Map: A Guide for Researchers

Contents

• Introduction to the Spatial Mapping Protocol
• Core Data Sources and System Architecture
• Protocol for Layer Configuration and Variable Control
• Temporal Analysis and Historical Data Integration
• Data Extraction and GIS Export Procedures
• Methodological Scope and System Limitations
• Application: Tracking a Boreal Forest Fire Complex
• Summary of Research Protocols

Introduction to the Spatial Mapping Protocol

The Alaska Wildfires Web Map exists for one reason: to put unmanipulated spatial data in the hands of people who need to analyze it themselves. Environmental scientists tracking active fire fronts and climate researchers reconstructing burn histories do not need another pre-rendered dashboard. They need vector data they can pull apart, reproject, and run through their own models.

That distinction shaped how we built this protocol. Rather than steering you toward polished dashboard views, the workflow below focuses on raw spatial extraction. You query the data, you export it, you verify it independently.

The map serves two broad tracking tasks. The first is monitoring active thermal fronts in something close to real time. The second is reconstructing historical burn scars across a window spanning 1990 to 2022, long enough to support meaningful longitudinal study in Alaskan ecoregions.

The goal here is a replicable protocol. Follow the steps, document your parameters, and another researcher should be able to reproduce your extraction without guessing at your settings. One operational constraint to note up front: extraction queries are limited to roughly 50,000 vertices per request, which forces deliberate spatial subsetting on large complexes.

Core Data Sources and System Architecture

The thermal layer draws primarily on VIIRS telemetry. We considered a MODIS-first architecture and rejected it. The deciding factor was spatial resolution: at 375 meters, VIIRS captures the fragmented, broken fire lines typical of boreal terrain far better than the coarser MODIS product. For fragmented burn geometry, that resolution gap matters more than legacy continuity.

Beyond the satellite feeds, the map aggregates incident data from the Alaska Interagency Coordination Center (AICC) along with other federal and state monitoring systems. AICC perimeters provide the ground-coordinated context that pure thermal detection lacks.

The detection-to-visualization pipeline

Data moves through a defined sequence. A polar-orbiting satellite detects a thermal anomaly, the signal is processed into a geolocated point, and that point is ingested into the web map. The temporal latency from overpass to visualization runs roughly 2 to 4 hours.

That latency window is not a defect. It reflects the real cost of processing and quality control between detection and display, and any researcher reporting near-real-time positions needs to state it explicitly.

Protocol for Layer Configuration and Variable Control

Isolating variables is the core analytical skill here. The map separates active thermal hotspots from contained perimeters, and you almost never want both at full opacity simultaneously.

1. Disable the contained-perimeter layer to expose only active thermal detections.
2. Confirm the thermal layer is drawing from VIIRS rather than a blended source.
3. Re-enable contained perimeters at reduced opacity once the active front is established.

Overlaying meteorological data

Wind vectors and smoke dispersion models add interpretive power, but they also clutter the canvas. We set the default layer hierarchy to place meteorological vectors below the active thermal hotspots. This prevents the wind field from occluding the primary fire front during dense multi-layer analysis.

These meteorological layers refresh every 6 to 8 hours. If you are correlating a wind shift with a perimeter change, check the timestamp on the vector layer before drawing conclusions; the wind field you see may predate the thermal detection beside it.

Calibrating thermal thresholds

Visual thresholds distinguish smoldering from active crown fire. The calibration sits at about 300 Kelvin for smoldering activity and 400 Kelvin for active crown fires. Set these deliberately. A threshold left at the smoldering floor will flood high-intensity complexes with low-confidence noise.

Temporal Analysis and Historical Data Integration

Longitudinal work begins with the historical perimeter layers. Activate them, set your date range, and the map renders past burn footprints against current conditions.

The time-slider animates fire progression, but it does not slide continuously. It steps in discrete 12-hour increments. We designed it this way to align the animation directly with polar-orbiting satellite overpass schedules; a continuous slider would imply fire positions between overpasses that the satellites never actually observed.

Comparing burn severity against fire return intervals

To compare current severity with historical baselines, you load the fire return interval layer for your ecoregion. In the Yukon Flats, those intervals are calculated over a 15 to 40 year span. That range becomes your reference frame for judging whether a current burn is anomalous or simply on schedule.

Fire return intervals in the tundra regions of the North Slope differ drastically from the boreal forests of the Interior. Loading a single statewide baseline will mislead you; each region requires its own distinct historical layer.

One performance constraint shapes long studies: longitudinal animation rendering is capped at 90-day intervals. Beyond that, browser memory becomes the bottleneck. Segment longer studies into stitched 90-day blocks.

Data Extraction and GIS Export Procedures

Extraction is where the protocol earns its replicability. The web interface exports to both Shapefile and GeoJSON.

1. Define your spatial subset, respecting the vertex query ceiling of roughly 50,000.
2. Select the output format. GeoJSON for web pipelines, Shapefile for legacy desktop workflows.
3. Confirm the export stays under the 250 MB per-query file size limit.
4. Download and verify the coordinate reference system before importing.

Coordinate reference system integrity

The default export CRS is EPSG:3338, Alaska Albers. We standardized on it to minimize spatial distortion across the state's enormous longitudinal spread. During GeoJSON to Shapefile conversion, the coordinate transformation tolerance is held at about 0.1 meters.

Risk Factor: Exporting in WGS 84 (EPSG:4326) without reprojecting to Alaska Albers produces severe area calculation errors for high-latitude burn scars. At these latitudes, geographic coordinates badly distort area. Always confirm the projection before computing burned hectares.

Importing into local GIS

Once exported, the datasets drop into QGIS or ArcGIS for advanced geoprocessing. Define the CRS on import rather than letting the software guess. If your environment defaults to a geographic CRS, set it to EPSG:3338 manually before running any area or distance calculation.

Methodological Scope and System Limitations

Satellite-based thermal detection has hard physical limits, and honest analysis names them.

Cloud cover and dense smoke canopies block the optical path. The map flags data when optical depth exceeds about 0.4, signaling that detections in that area are unreliable rather than absent.

Critical Insight: Thermal detection algorithms struggle to penetrate dense smoke canopies. Sub-canopy smoldering in black spruce forests may go entirely unrecorded until the fire breaches the crown. Absence of a detection is not evidence of absence of fire.

Spatial resolution imposes a second constraint. At 375 meters, VIIRS resolves fragmented or highly localized burn areas only coarsely. Small spot fires below the pixel footprint blend into surrounding terrain.

Temporal blind spots

Consecutive polar-orbiting passes leave gaps of 4 to 6 hours. Fire behavior during those windows is unobserved. We made a deliberate methodological choice not to interpolate spread across these gaps. Interpolation would manufacture data the satellites never collected, so the map flags the gap instead. For peer-reviewed work, that integrity matters more than a visually continuous animation.

Application: Tracking a Boreal Forest Fire Complex

Consider a multi-ignition boreal complex tracked over a 14 to 21 day burn period. The complex begins as several distinct ignitions that gradually merge into a single perimeter, which is exactly why it makes a useful demonstration of the full protocol.

Executing the protocol in sequence

The workflow runs in order. First, configure layers: isolate the VIIRS thermal front, set thresholds at 300 and 400 Kelvin, and drop the wind vectors beneath the hotspots. Next, drive the time-slider in 12-hour steps to watch the separate ignitions converge. Then extract the merged perimeter in EPSG:3338, subsetting to stay under the vertex ceiling.

Each distinct ignition gets its own extraction before the merge, then a combined extraction afterward. That sequence preserves the analytical record of how the complex assembled itself.

Synthesizing with field observations

The exported spatial data becomes one input among several. Researchers overlay the burn footprint against permafrost data to assess degradation risk at active layer depths of about 0.5 to 1.2 meters. Where a high-severity burn coincides with shallow active layers, the degradation risk concentrates.

Field observations close the loop. Satellite perimeters tell you where the fire was detected; ground crews tell you what actually burned beneath the canopy.

Summary of Research Protocols

Accuracy and replicability depend on a few non-negotiable steps. Document your thermal thresholds. Record your export CRS. State your latency and overpass-gap windows in any published analysis.

Understanding the system's limitations is not optional housekeeping; it is a precondition for valid spatial analysis. A burn scar area computed in the wrong projection, or a perimeter reported without its temporal blind spots, undermines everything downstream.

Recommendation: Cross-reference satellite telemetry with ground-truth reports before finalizing any analysis. Validation windows run roughly 48 to 72 hours post-detection, and community reports typically lag satellite detection by 12 to 24 hours. Build that lag into your timeline rather than treating field reports as immediately available.

Integrating web map data with on-the-ground Alaskan community observations remains the strongest path to validated perimeters. The satellite sees heat; the community sees what the heat means on the land. Both belong in a rigorous analysis.