28 Biodiversity Mapping Through Drone-Based Multispectral Sensors
Practical Module | International Summer Course 2026
This module introduces participants to drone-based remote sensing for biodiversity monitoring. Through supervised field activities and structured data analysis exercises, participants gain practical experience in aerial data acquisition, photogrammetric processing, and ecological interpretation of drone-derived spatial products. The module is designed to help participants understand both the technical workflow of drone mapping and the scientific considerations behind sensor selection, survey design, and spatial analysis.
28.1 Learning Objectives
By the end of this module, participants should be able to:
- Understand fundamental principles of drone-based remote sensing for biodiversity.
- Identify different drone platforms and sensor types used in ecological surveys.
- Apply basic flight planning principles and Ground Control Point (GCP) procedures.
- Operate a drone under instructor supervision during field missions.
- Process drone imagery to generate orthomosaics, NDVI maps, and surface models.
- Interpret ecological patterns from multispectral and LiDAR data products.
- Connect drone-derived spatial information to conservation and environmental policy questions.
28.2 Introduction to Drone Mapping for Biodiversity
Remote sensing is the process of obtaining information about Earth’s surface without physically touching or measuring it on the ground. Information is collected using sensors mounted on satellites, aircraft, or drones. These sensors detect reflected sunlight, emitted heat, or laser signals from vegetation, soil, water, and other surface features.
In biodiversity and conservation studies, remote sensing allows researchers to observe ecosystems efficiently across large or difficult-to-access areas. Using drones, scientists can monitor forests, wetlands, coral reefs, grasslands, wildlife habitats, and environmental disturbances with high spatial detail and flexible survey timing.
28.2.1 Why Use Drones for Mapping?
Conventional biodiversity surveys often require extensive fieldwork, long observation periods, and difficult access to remote areas. In many ecosystems, ground surveys alone may not provide sufficient spatial coverage or temporal frequency. Drone mapping helps overcome these limitations by enabling rapid and detailed aerial observation.
Drones provide several advantages for conservation applications:
- High-resolution spatial data
- Flexible flight scheduling
- Low operational cost compared to aircraft surveys
- Ability to access hazardous or remote environments
- Repeatable monitoring through time
- Reduced disturbance to sensitive ecosystems when properly operated
Drone mapping is increasingly used for forest canopy monitoring, habitat mapping, mangrove assessment, wildlife habitat analysis, peatland monitoring, coral reef mapping, invasive species detection, environmental restoration monitoring, and fire and post-disturbance assessment.
28.3 Drone Platforms and Sensor Types
28.3.1 Platform Types
Drone mapping uses two main types of platforms depending on the survey objective and area coverage needed.
Multirotor platforms (quadcopters, hexacopters) are capable of vertical takeoff and landing, hovering, and maneuvering in confined spaces. They are well-suited for detailed surveys of small to medium areas and can carry a variety of sensors. Their main limitation is relatively short flight endurance per battery charge.
Fixed-wing platforms generate lift through rigid wings and can cover substantially larger areas per flight. They require a clear launch and landing area and cannot hover. They are preferred for large-area landscape surveys.
28.3.2 Sensor Types
Depending on the survey objective, the drone may carry different sensor types:
- RGB cameras capture visible-light imagery for habitat documentation, land cover mapping, and 3D surface reconstruction.
- Multispectral sensors measure reflected energy including near-infrared wavelengths, enabling calculation of vegetation indices such as NDVI for vegetation health monitoring.
- Thermal sensors detect surface temperature variation, used for wildlife detection, microclimate mapping, and fire monitoring.
- LiDAR sensors emit laser pulses to measure 3D structure including canopy height, forest structure, and terrain beneath vegetation.
28.4 Survey Planning
28.4.1 Flight Parameters
Careful survey planning ensures that collected imagery is scientifically useful, spatially accurate, and appropriate to the ecological objective.
Flight altitude determines spatial resolution and coverage area. Lower altitude produces finer spatial detail but reduces area coverage per flight. The appropriate altitude depends on the ecological question and the level of detail needed.
Image overlap refers to the proportion of shared area between adjacent photographs. Typical ecological surveys use a front overlap of 70-85% and a side overlap of 60-80%. Higher overlap improves reconstruction quality but increases the number of images and processing time.
Ground Control Points (GCPs) are physical markers placed at known geographic positions within the survey area, measured with GNSS receivers. GCPs anchor the photogrammetric reconstruction to real-world coordinates and improve positional accuracy.
28.4.2 Environmental and Regulatory Considerations
Weather conditions strongly influence flight safety and data quality. Strong winds reduce stability, while rain or heavy cloud cover reduces image quality. Thermal surveys are often conducted early in the morning to maximize thermal contrast.
Drone operations must comply with local aviation regulations, institutional policies, and protected-area management rules. In Indonesia, flights require authorization from the civil aviation authority and compliance with altitude restrictions.
28.5 Fieldwork Locations – ISC Biotrop 2026
This module is conducted at two contrasting field sites selected to represent distinct ecosystem types and biodiversity conditions.
Bayat, Klaten is a terrestrial site characterised by a mosaic of agricultural land, secondary vegetation, and remnant natural patches. This location allows participants to observe how drone imagery captures land cover heterogeneity, vegetation condition variation, and habitat patch boundaries in a human-modified landscape.
Pantai Porok, Gunungkidul is an intertidal coastal site supporting marine biodiversity communities on the intertidal flat. This location presents mapping challenges including highly reflective water surfaces, rapidly changing tidal conditions, and substrate types ranging from bare rock to algal mat and seagrass.
28.6 Photogrammetric Processing
Photogrammetric processing reconstructs three-dimensional spatial products from overlapping drone photographs.
The processing workflow typically involves:
- Image import and quality assessment – reviewing imagery for sharpness, exposure, and coverage.
- Image alignment – identifying matching features across overlapping images to estimate camera positions.
- Georeferencing – registering GCP coordinates to anchor the model to real-world coordinates.
- Dense point cloud generation – producing millions of 3D points representing the surveyed surface.
- Digital Surface Model (DSM) and Digital Terrain Model (DTM) – generating elevation grids from point cloud data.
- Canopy Height Model (CHM) – derived by subtracting DTM from DSM:
CHM = DSM - DTM. - Orthomosaic generation – producing a geometrically corrected, mosaicked image product.
Multispectral data requires radiometric calibration before vegetation indices can be calculated. A calibration panel with known reflectance values is imaged before and after each flight.
28.7 Exercises
28.7.1 Exercise 1 – Supervised Drone Operation
This exercise gives participants supervised practical experience in drone operation. Participants observe or participate in flight preparation, pre-flight inspection, mission execution, and data transfer under instructor supervision.
Participants observe:
- Pre-flight equipment checks (battery, propellers, GNSS signal, sensor calibration)
- GCP placement and measurement
- Automated flight mission execution
- Post-flight data management
All flight activities are conducted by authorized instructors or certified operators. Participants are assigned team roles including Mission Supervisor, UAV Pilot, Visual Observer, GCP Team, and Data Manager.
28.7.2 Exercise 2 – NDVI Analysis
This exercise introduces participants to vegetation condition analysis using the Normalized Difference Vegetation Index (NDVI) derived from drone multispectral data.
NDVI uses the near-infrared and red spectral bands:
\[\text{NDVI} = \frac{NIR - Red}{NIR + Red}\]
Participants observe NDVI maps showing:
- Dense healthy canopy
- Canopy gaps
- Understory vegetation
- Stressed vegetation patches
- And edge effects near built environments
Objectives: Participants should calculate NDVI, visualise vegetation health patterns, identify spatial variation in vegetation condition, and interpret ecological meaning from spectral patterns.
Protocol:
- Step 1 – Import multispectral dataset into GIS or remote sensing software
- Step 2 – Identify the Red band and Near-Infrared (NIR) band
- Step 3 – Calculate NDVI using the formula above
- Step 4 – Generate a colour-coded NDVI map
- Step 5 – Interpret ecological patterns (healthy canopy zones, low-vegetation areas, understory patterns, disturbed regions)
Recommended Free Software: QGIS, SNAP (ESA), WhiteboxTools
28.7.3 Exercise 3 – LiDAR Point Cloud Visualisation
This exercise introduces participants to three-dimensional forest visualisation using LiDAR point cloud data. Participants explore how LiDAR captures structural information from both canopy and terrain surfaces.
Participants observe:
- Canopy height variation
- Understory visibility
- Terrain beneath vegetation
- And structural ecosystem complexity
The exercise demonstrates how LiDAR differs from standard aerial imagery by capturing forest structure in 3D.
Objectives: Participants should visualise point cloud datasets, distinguish canopy and terrain points, and interpret forest structural complexity.
Recommended Free Software: CloudCompare, LAStools, QGIS
28.7.4 Exercise 4 – Canopy Height and Forest Structure Analysis
Using DSM and DTM products, participants generate a Canopy Height Model (CHM) to estimate vegetation height and analyse forest structure.
Canopy height provides ecological information related to:
- Vegetation maturity
- Biomass
- Habitat complexity
- And ecosystem condition
Participants compare canopy variation across the forest and identify tall canopy regions, open gaps, understory zones, and structural heterogeneity.
Protocol:
- Step 1 – Import DSM and DTM into GIS software
- Step 2 – Calculate CHM:
CHM = DSM - DTM - Step 3 – Generate colour-coded canopy height maps
- Step 4 – Interpret ecological structure (mature canopy, canopy gaps, understory complexity, habitat variation)
Recommended Free Software: QGIS, SAGA GIS, WhiteboxTools
28.7.5 Exercise 5 – Extracting Ecological Insights and Policy Relevance
The final exercise focuses on transforming spatial analysis into ecological interpretation and conservation-related recommendations. Participants analyse the generated products collectively and discuss how drone-derived information may support environmental policy and ecosystem management.
Potential discussion topics include:
- Identifying vegetation stress areas
- Detecting habitat fragmentation
- Evaluating canopy connectivity
- Recognising restoration priorities
- Monitoring urban biodiversity
- And assessing ecosystem resilience
Participants should understand that drone mapping products are not only technical outputs, but also decision-support tools for environmental management and conservation planning.
For example:
- Areas with low NDVI may indicate vegetation stress requiring ecological intervention
- A fragmented canopy structure may influence wildlife movement
- Understory visibility may indicate habitat openness
- And terrain analysis may support hydrological management
Example Policy-Relevant Questions:
- Which areas should be prioritised for restoration?
- Are there signs of ecosystem degradation?
- How can canopy connectivity support biodiversity conservation?
- Which areas are vulnerable to environmental disturbance?
- How might campus development affect ecological structure?
28.8 Summary
The exercises presented in this module are designed to help participants understand how drone-derived multispectral and LiDAR datasets can support biodiversity and conservation analysis. Through supervised drone operation, vegetation analysis, LiDAR visualisation, canopy structure assessment, and ecological interpretation, participants learn how remote sensing products can be transformed into meaningful environmental information.
The Faculty of Biology UGM forest dataset provides a realistic ecological case study that allows participants to explore tropical forest structure, understory vegetation, and habitat complexity using modern drone mapping approaches. By connecting spatial analysis with ecological reasoning and policy relevance, the exercises demonstrate how drone-based remote sensing can contribute to evidence-based conservation and environmental management.
28.9 Full Workshop Module
The complete practical module document is embedded below for reference and offline reading.