10 GSI Datasets Useful for Heat Map Analysis

What matters in heat map analysis is not coloring itself but what you quantify and which geographic information you overlay to give it meaning. The Geospatial Information Authority of Japan (GSI) publishes a variety of information for understanding Japan’s territory—topographic maps, aerial photographs, elevation, terrain classifications, disaster information, etc.—on GSI Maps and other services, and supports overlaying and 3D display. For this reason, GSI data can serve as a foundation that elevates a heat map from a mere visual to geographic information usable for decision-making.

Many practitioners who search for “heat map GSI” wonder which data to choose: should they look at elevation, aerial photographs, or include disaster and land-condition data? The short answer is that it’s important to use different data according to your objective. The optimal GSI dataset depends on whether you want to see elevation differences, flood susceptibility, vegetation bias, or historical terrain change. This article narrows down GSI datasets that are truly useful for heat map analysis to 10 types and organizes their practical uses clearly.

Table of contents

• Why GSI data are used in heat map analysis

• Fundamental Geospatial Data (basic items)

• Fundamental Geospatial Data Numerical Elevation Model

• Elevation tiles

• Digital National Land Basic Map (orthoimages)

• Aerial photographs by year

• Land Condition Map

• Topographic classification (vector tiles)

• Flood-control topographic classification map

• Nationwide Vegetation Index Data (250 m (820.2 ft))

• Point cloud data

• How to combine datasets in practice without getting lost

• Common pitfalls in heat map analysis

• Summary

Why GSI data are used in heat map analysis

A heat map expresses the intensity and distribution of numbers with color, but in the geospatial world simple color gradation alone is often insufficient. For example, a red high-temperature area may be due to pavement, low-lying stagnation, surrounding land use, or terrain effects—without background map information you can misread the cause. The strength of GSI data lies in the ease of handling elevation, aerial photographs, terrain classification, and disaster-related information under the same geographic coordinate framework. Ease of alignment greatly affects the reproducibility and explanatory power of heat map analysis.

Another important point is that GSI data are layered from broad-coverage datasets to detailed analysis-ready datasets. You can first grasp trends on a web map and then dig deeper with a numerical elevation model or point cloud data, making staged analysis easy. This is useful in many practical phases: planning, desktop study before field confirmation, report preparation, resident briefings, pre-construction terrain checks, etc. A key to successful heat map analysis is not trying to complete everything with a single dataset from the start. It is important to combine overview data with data that help read causes.

1. Fundamental Geospatial Data (basic items)

The first thing to grasp is the basic items of Fundamental Geospatial Data. GSI positions Fundamental Geospatial Data as “information that serves as the positional basis in electronic maps,” and explains that if various entities build geospatial information with the same positional standard, they can be correctly linked and overlaid. In heat map analysis, this property of “correctly overlapping” is extremely important. No matter how attractive the color distribution, if it is misaligned with baseline information such as roads, waterlines, administrative boundaries, or building footprints, the credibility of the analysis drops sharply.

Fundamental Geospatial Data (basic items) are, in a sense, data suited for building the foundation of a heat map. When visualizing population, incident reports, inspection results, movement history, accident distributions, or photo capture locations with color, you ultimately need to explain which road, which river vicinity, or which parcel is involved. With well-prepared base data, you can more easily return the meaning of colors to their geographic context. In practice, these data are more effective for setting the clip range, mask processing, organizing aggregation units, and checking consistency with background maps than for the analysis itself.

2. Fundamental Geospatial Data Numerical Elevation Model

One of the most frequently used GSI datasets in heat map analysis is the Numerical Elevation Model of the Fundamental Geospatial Data. This mesh-based elevation dataset allows you to treat surface elevation differences as numerical values, directly supporting elevation heat maps, slope distribution, valley extraction, examination of areas where rainwater accumulates, and comparisons before and after land development. GSI provides the numerical elevation model on the Fundamental Geospatial Data site and shows its maintenance and update status. As a starting point for heat map analysis, checking the numerical elevation model first is very reasonable.

A practical advantage is that it makes it easy to quantify slight terrain differences that are not visually obvious. In urban areas, differences of tens of centimeters to several meters can affect drainage or flooding behavior, and on developed sites or slopes, simply understanding slope continuity increases available judgment information. When converting to a heat map, it is effective not only to color by absolute elevation but also by difference from surrounding area, difference from a specified reference surface, or by converting to slope amount. Deciding what value to color before worrying about flashy colors greatly affects analysis quality.

3. Elevation tiles

When you want to quickly try an elevation-based heat map, elevation tiles used in GSI Maps are convenient. GSI explains that elevation tiles are prepared using the same tile coordinates and pixel coordinates as map tiles and are available in text and PNG formats. This makes them suitable for initial investigations of elevation on web maps. Because they are easy to overlay with maps without large preprocessing, they are very handy for getting a quick sense of trends.

The value of elevation tiles lies in the speed of prototyping. You can quickly survey a wide area to see where elevations rise or where lowlands spread and where color transitions are abrupt. In the planning phase this speed matters. In stakeholder meetings, showing a color distribution derived from elevation tiles to form hypotheses and then moving to detailed numerical elevation models helps progress discussions. However, elevation tiles are excellent as an entry for “quick viewing,” and for final accuracy checks or cross-sectional examination you must compare with higher-fidelity elevation models or point cloud data.

4. Digital National Land Basic Map (orthoimages)

Orthoimages from the Digital National Land Basic Map are powerful for linking color distributions to real-world conditions. GSI states that it converts captured aerial photographs into positionally accurate images and provides them as easy-to-use image information with correct positional information as the Digital National Land Basic Map (orthoimages). In other words, orthoimages are not “photographs as-is” but images that can be correctly overlaid with other geospatial information. They are very helpful for confirming the cause of colors in heat map analysis.

For example, when you check a high-temperature area, you may find it is a parking lot, a rooftop, bare developed land, or a slope before protection—cases where viewing the photograph is necessary to understand. Conversely, if vegetation should be high but the heat map shows a high value, you may be led to review analysis conditions such as seasonal differences, shadows, or aggregation mismatches. Orthoimages are not data that merely explain the heat map legend; they reduce misinterpretation of heat maps. Always confirm the current situation with photographs before interpreting colors—this extra step elevates desktop analysis to practical level.

5. Aerial photographs by year

For heat maps that involve change analysis, aerial photographs by year are very effective. GSI Maps provides aerial photographs from before World War II to the present in various years, and you can display them side by side or overlay them for comparison. GSI also repeatedly photographed aerial photos from the late 1940s to the present and makes them available for viewing and download via the “Map and Aerial Photograph Viewing Service.” Being able to check past terrain and land use is a major advantage when investigating the causes of current color distributions.

Heat maps show current distributions well but do not automatically tell you why a place became like that. A site that is now residential may once have been a back swamp or an old river channel; a currently stable-looking filled site may have been a valley in the past. Overlaying aerial photographs by year lets you read the history of the land. In practice, avoid rushing to conclusions from a single-year heat map; at minimum, check past photographs once. Understanding terrain change and land use transition history deepens the meaning you assign to colors.

6. Land Condition Map

If you want to read heat maps with disaster prevention and land use perspectives, the Land Condition Map is indispensable. According to GSI, the Land Condition Map mainly indicates terrain classification based on results of land condition surveys conducted to provide fundamental materials related to natural land conditions necessary for formulating plans such as disaster prevention measures, land use, land conservation, and regional development. In other words, this is information for reading the nature of land, not merely a background map.

The Land Condition Map is useful in heat map analysis because it helps you recognize that “the same color can mean different things.” For example, a blue area indicating low elevation could be interpreted differently depending on whether it is behind a natural levee, an artificially altered lowland, or reclaimed land. Overlaying the Land Condition Map allows you to consider land formation, not just relative values. Especially for disaster-related heat maps, site evaluations, facility placement, and early-stage planning of land development, referencing the Land Condition Map early reduces the risk of later significant oversights.

7. Topographic classification (vector tiles)

Topographic classification vector tiles available on GSI Maps are easy to explain to field staff and non-specialists. GSI explains that the “Topographic classification (vector tile provision experiment)” displays a colored topographic classification map on GSI Maps, and clicking shows the land’s formation and natural disaster risks. Because it handles attributes more easily than ordinary image maps, it pairs well with heat map overlays and explanatory materials.

A strength of this data is that it makes it easy to organize the meaning behind colors in terms closer to practical judgment rather than just technical terminology. For example, if abnormal values are continuous in an area, seeing whether that area is along an old river channel, an alluvial fan, or a filled land changes what you need to check next. Topographic classification (vector tiles) is not strict numerical analysis data but a powerful auxiliary line for hypothesis formation and explanation. It helps you answer in terrain terms when someone asks in a meeting, “What is this red band?”

8. Flood-control topographic classification map

If your heat map analysis focuses on flood damage, internal water problems, or lowland vulnerability, the flood-control topographic classification map is highly practical. GSI explains that this map mainly targets lowlands formed by river or sea actions, which are closely related to flood control, and classifies terrain elements prone to flooding in detail. It also classifies artificially modified terrain such as filled land, reclaimed land, and cut areas. This information pairs extremely well with water-related heat maps.

For example, when showing distributions of flood history, poor drainage, inundation reports, groundwater level, pavement damage, or liquefaction-related occurrences with color, overlaying the flood-control topographic classification map makes lowland-specific biases easier to see. Where a heat map alone might only show “higher values along rivers,” layering this map lets you determine whether it’s the edge of a natural levee, an old flow path, or reclaimed land. In practice, the important thing is accountability for results: being able to supplement the distribution with land formation explains why a place is high-risk, which is a major advantage.

9. Nationwide Vegetation Index Data (250 m (820.2 ft))

If you want to see vegetation bias or seasonal change with heat maps, the Nationwide Vegetation Index Data (250 m (820.2 ft)) is a strong candidate. The GSI tile list includes the Nationwide Vegetation Index Data (250 m (820.2 ft)), and GSI provides a vegetation index data service. Vegetation indices are generally suitable for grasping vegetation vigor and distribution trends at a wide area, useful for urban green coverage, seasonal changes in mountainous areas, and capturing regional environmental tendencies. This is a representative dataset that pairs well with heat maps.

However, this dataset corresponds to approximately a 250 m mesh and is intended for wide-area use. Therefore, it is not suitable for block- or site-level detailed assessments. Misunderstanding this can lead to applying coarse data to fine-grained on-site decisions. Conversely, it is very strong for broad-area overviews. If you want to see bias in greenery across an entire city, grasp vegetation density across a watershed, or roughly capture seasonal differences, it is practically useful. Use it as a layer for grasping large-scale trends and do not confuse it with high-resolution photos or ground-surface data.

10. Point cloud data

If you move to more detailed three-dimensional heat map analysis, point cloud data are powerful. GSI describes point cloud data as collections of measured points obtained by airborne laser surveys and states that they include surface heights of not only the ground but also buildings and vegetation, color information, reflectance intensity, and simple classifications. They also note that acquisition density is 4 points or more per 1 m^2, vertical accuracy is about 25 cm (9.8 in), and one file ranges from several hundred MB to several GB. Point cloud data add great value when you want to capture fine surface undulations and the influence of structures that mesh elevation alone cannot grasp.

Using point cloud data for heat maps makes it easier to color and express fine spatial differences such as local slope deformations, unevenness of embankments, micro-topography of levees and road shoulders, level differences around buildings, and vegetation cover bias. Especially for construction management, infrastructure inspection, earthwork quantity comparisons, and deformation detection, difference heat maps derived from point clouds are powerful. On the other hand, data volumes are large and preprocessing and display loads tend to be high. Therefore, use elevation models for broad overviews and point clouds for local inspections to make operations manageable.

How to combine datasets in practice without getting lost

Having reviewed 10 types, in practice it is overwhelmingly easier to combine them according to purpose rather than choosing just one. For example, if you want to see ground elevation differences or drainage bias, center your approach on the Fundamental Geospatial Data numerical elevation model, use elevation tiles to quickly check overall trends, and verify the current situation with Digital National Land Basic Map orthoimages. If you then overlay the Land Condition Map or the flood-control topographic classification map, you can distinguish whether it is merely lowland or lowland strongly related to water disasters.

If you want to examine urban environment or green bias, start with the Nationwide Vegetation Index Data (250 m (820.2 ft)) to grasp broad trends and verify current status and history with orthoimages and aerial photographs by year. This lets you not only find places with little greenery but also read when the reduction occurred and whether it was due to development or paving, or related to surrounding terrain. You can make heat maps alone, but if you want to read causes, always attach a different type of data. That is the iron rule.

For field-sensitive tasks like construction or infrastructure management, center on point cloud data or detailed elevation data, align positions with Fundamental Geospatial Data, and confirm appearance with aerial photographs. Grasp trends with GSI’s broad-area data and finalize with field measurements or construction data—this two-stage approach links desktop study to field judgment. The important thing is not to seek a single all-purpose map from the start. Think in three layers: positional base data, data that creates values, and images/classification maps that assist interpretation; this reduces indecision in data selection.

Common pitfalls in heat map analysis

When using GSI data, the first thing to watch is differences in coordinate systems and elevation results. The Fundamental Geospatial Data Download Service indicates that basic items and numerical elevation models provided on or after July 31, 2025 reflect elevation revisions and that the coordinate reference system changed from JGD2011 to JGD2024. Also, GSI Maps help indicates that on March 31, 2026 elevation tiles were updated to the 2024 geodetic results after the elevation revision. When mixing past data or internally held datasets, do not overlook these differences.

Next, be careful about mismatches in resolution and intended use. The Nationwide Vegetation Index Data (250 m (820.2 ft)) is for wide-area trends, point cloud data are high-resolution but heavy, and elevation tiles are suitable for initial checks but may not provide final answers for detailed verification. That is, it is about fit-for-purpose rather than superiority. Do not make fine judgments with coarse data, and do not force high-resolution data into broad-area explanations from the outset. Do not judge importance only by color intensity; check the source data’s granularity, acquisition date, purpose, and update status before evaluation.

Also be aware that heat map colors strongly influence viewers, and presentation can dominate. Simplifications like “red is dangerous, blue is safe” can cause misunderstanding in practice. The important thing is to make clear what value is being colored, whether that value is measured, estimated, or a classification. GSI data are a reliable foundation, but they do not automatically guarantee the meaning of your heat map. Only when you clarify the legend, units, period, and comparison standards can a figure fulfill explanatory responsibility.

Summary

When choosing GSI data useful for heat map analysis, first decide “what value do you want to color,” and then select data that support “how to interpret that color.” For positional alignment, use Fundamental Geospatial Data; for quantifying terrain, use numerical elevation models and elevation tiles; for confirming current conditions, use Digital National Land Basic Map (orthoimages) and aerial photographs by year; to view land formation and disaster relationships, use Land Condition Map, topographic classification, and flood-control topographic classification map; for environmental trends, use Nationwide Vegetation Index Data; and for detailed 3D analysis, use point cloud data. No single dataset is万能; combining datasets according to purpose makes a heat map practically usable.

After grasping trends with GSI’s wide-area data, what matters in the field is connecting those trends to precise field measurements without shift. If you want to overlay points, photos, as-built, and equipment locations with high accuracy and translate desktop heat map analysis into field decisions, using LRTK is highly compatible. LRTK, as an iPhone-mounted GNSS high-precision positioning device, is a practical choice when you need high-precision field positioning. Use GSI data for broad understanding and LRTK to accurately pin down the field. This workflow makes analysis more than “visualization”: it becomes a practical tool from simple surveying to verification, recording, and sharing.