GIS survey and field data collection: how GIS works, data types and accuracy
Key takeaways
- GIS means geographic information system: hardware, software, data, people and methods for storing, analysing and displaying located information.
- GIS data is either vector (points, lines, polygons with attributes) or raster (grids of cells such as imagery and elevation models).
- GeoPackage is a single-file, multi-layer open format; shapefiles still dominate exchange but have limits such as 10-character field names.
- Field accuracy tiers run from about 3–5 m for phones, 0.5–2 m for mapping-grade receivers to 1–3 cm for RTK.
- Quality assurance means checking known points, fix status, attributes, topology and metadata before data reaches the master database.
What is a GIS survey?
A GIS survey is a field survey whose purpose is to capture features and their descriptive attributes for a geographic information system (GIS), rather than to produce an engineering drawing or legal plan. Each recorded item, a hydrant, a manhole, a tree or a road sign, becomes a feature with a position, a geometry and a set of attributes such as material, condition and installation date.
GIS means geographic information system: a system for storing, analysing and displaying geographical data, meaning any data tied to a location on the Earth. A GIS survey map is the result of that collection, displayed as layers that can be queried, analysed and updated. GIS surveys are the field side of geospatial mapping, often shortened to geo mapping: whatever the office later does with location mapping and analysis depends on what crews record on site.
This article is a practical introduction to GIS for field crews: how the system works, the data types and file formats you will meet, and how accurate your collection needs to be. For a wider explanation of geospatial data and mapping, see our guide to geospatial data and GIS mapping.
What are the components of GIS?
The components of GIS are usually listed as five: hardware, software, data, people and methods. A GIS survey touches all five, and weak links in any one show up as bad data.
| Component | What it includes | In a GIS survey |
|---|---|---|
| Hardware | Servers, desktops, tablets, GNSS receivers | Field controller or rugged tablet plus GNSS rover |
| Software | Desktop GIS, web maps, field apps, databases | Data collection forms, sync, editing and analysis tools |
| Data | Vector layers, rasters, attribute tables, metadata | Features and attributes collected on site |
| People | GIS analysts, surveyors, field staff, managers | Crews trained on the schema and the equipment |
| Methods | Standards, procedures, QA rules | Data dictionary, accuracy targets, check routines |
Geographic information system tools range from desktop and server packages to open source GIS options and lightweight digital mapping tools in a web browser. The software choice matters less for a field team than an agreed data dictionary: the list of feature types, attributes and allowed values everyone collects to.
How does GIS work?
GIS works by storing every piece of information with a location in a known coordinate system, organising it in thematic layers, and linking each feature to a table of attributes. Because all layers share the same coordinate framework, the software can overlay them, measure, query and analyse relationships between them.
- Georeferencing. Every feature or image is tied to a coordinate reference system, identified by an EPSG code. Mixing systems without transformation is the most common cause of features appearing in the wrong place; see our guide to datums and coordinate systems.
- Layers. Roads, parcels, pipes, land cover and elevation are kept as separate GIS datasets that can be turned on, combined or analysed together.
- Attributes. Each feature has a row in a table: a valve's diameter, a parcel's owner, a tree's species.
- Spatial analysis. Spatial data analysis in GIS includes buffers ("all properties within 50 m of the main"), overlays, network tracing, density and terrain analysis.
- Mapping and visualization. Results are shown as maps, dashboards and 3D scenes; good map data visualization makes the pattern obvious to decision-makers.
That, in short, is how GIS works: it is much more than computerized mapping, because the map is just one view of a database that can answer location questions. The same data can drive web mapping applications, geospatial visualization in 3D, or GIS planning models for growth and infrastructure.
Vector vs raster: what types of GIS data are there?
GIS data comes in two main types: vector data, which represents discrete features as points, lines and polygons, and raster data, which represents continuous surfaces as a grid of cells. Field GIS surveys mostly produce vector data.
| Aspect | Vector | Raster |
|---|---|---|
| Structure | Points, lines, polygons with coordinates | Grid of cells (pixels), each with a value |
| Examples | Hydrants, pipes, roads, parcel boundaries | Aerial imagery, elevation models, land cover |
| Attributes | Rich table per feature | One or a few values per cell |
| Precision | As precise as the measured coordinates | Limited by cell size (resolution) |
| Typical source | GNSS field survey, digitising, CAD | Satellites, drones, lidar, scanned maps |
| Common formats | Shapefile, GeoPackage, GeoJSON, KML | GeoTIFF, JPEG 2000, raster GeoPackage tiles |
Most projects combine both. A typical land mapping project overlays surveyed vector features on an orthophoto or a terrain model, then uses spatial mapping tools to derive slopes, catchments or visibility.
Which GIS file formats should field data use?
Use GeoPackage for most new field projects, keep shapefile for exchange with systems that require it, and use GeoJSON or KML for web display and sharing. Choose the format before collection starts so attribute names and types survive the round trip.
| Format | What it is | Strengths | Limits |
|---|---|---|---|
| Shapefile | Set of files: .shp geometry, .shx index, .dbf attributes, .prj projection | Universally supported | One geometry type per file, 10-character field names, about 2 GB per file, easy to lose a component |
| GeoPackage (.gpkg) | Open OGC standard based on a single SQLite file | Many layers in one file, vector and raster, long field names | Older software may not read it |
| GeoJSON | Text format for features (RFC 7946) | Web maps, APIs, readable text | Longitude/latitude in WGS84 only; large files are slow |
| KML / KMZ | XML format, OGC standard, KMZ is zipped | Easy viewing in virtual globes, sharing with clients | Built for display; limited attribute and projection handling |
| CSV with coordinates | Plain table with X, Y, Z columns | Simple import from any controller | Points only; coordinate system must be documented separately |
One field tip: the attributes of a shapefile live in its shapefile database file (.dbf). If a client sends only the .shp, you have geometry without attributes, and without the .prj you have no coordinate system. Always zip and send all components together.
How accurate does a GIS survey need to be?
A GIS survey needs to be as accurate as the decisions made from it: a few metres is fine for a tree inventory, sub-metre for most asset mapping, and centimetres for underground utilities, parcel mapping GIS updates tied to survey control, and engineering design.
| Tier | Typical horizontal accuracy | Equipment | Suitable for |
|---|---|---|---|
| Consumer | About 3–5 m | Phone or tablet internal GPS | Inspections, rough inventories, citizen reporting |
| Mapping grade | About 0.5–2 m | GNSS with SBAS or similar corrections | Signs, trees, land cover, general asset mapping |
| High accuracy | About 10–30 cm | Multi-frequency GNSS, float or PPP solutions | Asset records where features are close together |
| Survey grade | About 1–3 cm | RTK GNSS with network or base corrections | Utilities, valves, kerbs, parcel control, as-builts |
For centimetre collection without a heavy pole kit, our P2 handheld GIS RTK rover weighs 199 g, connects to CORS directly over 4G and is rated at 8 mm + 1 ppm horizontal RTK accuracy. The Mini Network RTK Rover (409 g, 60° IMU tilt) streams centimetre positions to a phone or tablet running your field apps. Both require a local CORS account. For utility-specific requirements, see our guide to utility mapping with GNSS.
GIS survey workflow: how to collect field data step by step
A reliable GIS survey workflow moves from schema to field to quality check to publication. Skipping the planning steps is what produces data nobody trusts.
- Define the purpose and accuracy. Decide what decisions the data supports and which accuracy tier is needed for each feature type.
- Build the data dictionary. List feature types, geometry (point, line, polygon), attributes, domains (pick lists) and mandatory fields.
- Set the coordinate system. Match the client's master GIS, including datum, projection, geoid and EPSG code.
- Prepare devices and field apps. Whether you use a commercial GIS mapping app or open data collection programs, load forms, background maps and existing GIS data onto the controller or tablet, and configure the GNSS correction source.
- Check in on control. Measure a known point at the start of each day and record the difference.
- Collect features. Record position, attributes and a photo; store fix status, number of satellites and estimated accuracy with each feature as metadata.
- Sync and review daily. Upload to the office, review on a map, and fix gaps while the crew is still nearby.
- Quality-check and publish into the master database or deliver in the agreed format.
The tablet matters more than people expect: sunlight readability, battery life and sealing decide whether crews keep collecting in poor weather. Our T8 Pro rugged GNSS tablet is an 8-inch, sunlight-readable, IP67 Android tablet with a high-precision GNSS module, external antenna and 4G, so it can run GIS field apps standalone or with an RTK rover. Our rugged tablet buying guide covers what else to check.
How do you quality-check GIS survey data?
You quality-check GIS survey data by verifying positional accuracy against known points, filtering on the stored GNSS metadata, validating attributes and checking topology before the data is merged into the master dataset.
- Positional checks: compare daily check-point observations with published values; re-observe a sample of features (5–10 % is common) independently.
- Fix status filter: flag any feature recorded on float or autonomous when RTK was specified, or with an estimated accuracy above the project limit.
- Attribute validation: no blank mandatory fields, values within domains, consistent units.
- Topology: lines that should connect do, no duplicate points, polygons close and do not overlap where they should not.
- Metadata: collection date, device, crew, coordinate system and accuracy recorded for each layer.
Uses of GIS keep growing, from asset management and GIS city planning to agriculture, environmental monitoring and emergency response, and every one of those applications of GIS depends on field data that passes these checks. If you are equipping GIS crews, browse our GIS and utilities solutions or request a quote for handheld RTK rovers and rugged tablets matched to your accuracy tier.
Frequently asked questions
What does GIS mean?
GIS means geographic information system. It is a combination of hardware, software, data, people and methods used to capture, store, analyse and display information that is tied to locations on the Earth. Unlike a plain map, a GIS links every feature to a table of attributes and lets you query and analyse the relationships between layers.
How does a GIS system work?
A GIS system stores data as layers that share a common coordinate reference system. Each feature has geometry, such as a point, line or polygon, and a row of attributes. Because the layers are aligned, the software can overlay them, run spatial queries and analysis such as buffers or network tracing, and display the results as maps or dashboards.
What is the difference between a GIS survey and a land survey?
A land survey measures positions to establish boundaries, design topography or set out construction, often to legal or engineering standards. A GIS survey captures features and their attributes for an asset or planning database. GIS surveys can be survey-grade when RTK is used, but they focus on attributes and completeness as much as on positional precision.
What are geospatial technologies?
Geospatial technologies are the tools used to capture, manage and analyse location data. They include satellite positioning such as GNSS, remote sensing from satellites and aircraft, drones, lidar and laser scanning, total stations, and GIS software for mapping, analysis and visualisation. In practice, a GIS survey uses several of them together in one workflow.
Which file format is best for GIS data?
For most new projects GeoPackage is the best all-round format because it stores many vector and raster layers in a single open file with no short field-name limits. Shapefile remains the most widely accepted exchange format. GeoJSON suits web maps and APIs, while KML and KMZ suit sharing data for viewing in virtual globe applications.
Can you do a GIS survey with a phone?
Yes, for low-accuracy work. A phone's internal GPS is typically accurate to about three to five metres, which is enough for rough inventories and inspections. For sub-metre or centimetre work, connect the phone to an external GNSS receiver with corrections, such as a network RTK rover, and record fix status and accuracy with every feature.
What are some GIS map examples?
Common GIS map examples include utility network maps showing pipes and valves, parcel maps for property and tax records, zoning and land-use maps for planning, flood risk maps derived from elevation data, crop yield maps in agriculture, and asset maps of street lights, signs and trees. Each combines surveyed features with attribute data and background imagery.
Need equipment for this work?
Tell us the job, the country and the quantity. An engineer replies with a configuration and price within one business day. MOQ is one unit; OEM is available.
