MERIT DEM offers superior vertical accuracy (±2.668m RMSE) over other global models in diverse Malaysian terrains.

Why It Matters

The selection of an appropriate Digital Elevation Model (DEM) is critical for the accuracy of geospatial analyses in fields like hydrology, geomorphology, and environmental planning. This research provides empirical data to guide designers and researchers in choosing the most reliable DEM for their specific project needs, especially when working with diverse land cover types.

Key Finding

MERIT DEM is the most vertically accurate global model tested, with accuracy decreasing in forest areas. While improvised models offer slight improvements over SRTM, land cover type is a major factor influencing accuracy.

Key Findings

• MERIT DEM exhibited the best vertical accuracy with an RMSE of ±2.668m, outperforming SRTM (±5.666m), NASA (±3.656m), and EarthEnv (±5.948m) when compared to GNSS data.
• All tested DEMs showed higher accuracy in non-forest areas, followed by urban areas, and the lowest accuracy in forest areas, indicating land cover significantly impacts DEM performance.
• The improvised DEMs showed only slight improvements over the original SRTM DEM.

Research Evidence

Aim: To comprehensively evaluate the vertical accuracy of improvised global Digital Elevation Models (MERIT, NASADEM, EarthEnv) against high-resolution reference data and GNSS points across varied land cover types in Peninsular Malaysia.

Method: Comparative accuracy assessment

Procedure: The study compared the vertical accuracy of MERIT, NASADEM, EarthEnv, and SRTM DEMs using 7757 GNSS points and two high-resolution reference DEMs (TanDEM-X and airborne IFSAR). Accuracy was assessed over urban, non-forest, and forest land cover areas, and Root Mean Square Error (RMSE) was calculated.

Sample Size: 7757 GNSS points

Context: Geospatial data analysis, environmental modelling, land cover mapping

Design Principle

Validate geospatial data sources against project-specific accuracy requirements and local ground truth where possible.

How to Apply

When undertaking a design project that relies on elevation data (e.g., site suitability analysis, hydrological modelling, infrastructure planning), conduct a preliminary review of available global DEMs and select the one with the best-documented accuracy for the expected terrain and land cover conditions, prioritizing MERIT if high vertical accuracy is paramount.

Limitations

The study was conducted in a specific geographic region (northern Peninsular Malaysia), and findings may vary in different geographical and geological settings. The assessment was limited to vertical accuracy.

Student Guide (IB Design Technology)

Simple Explanation: This study found that the MERIT digital elevation model is the most accurate for measuring height compared to other free global models like SRTM, NASA, and EarthEnv. Its accuracy is best in open areas and worse in forests.

Why This Matters: Understanding the accuracy of different DEMs helps you select the best data for your design project, ensuring more reliable results for analyses like site planning or environmental impact assessments.

Critical Thinking: How might the differences in accuracy between these DEMs influence the outcome of a hydrological simulation or a landslide susceptibility analysis?

IA-Ready Paragraph: The selection of a Digital Elevation Model (DEM) is a critical step in many design projects. Research indicates that MERIT DEM offers superior vertical accuracy (RMSE ±2.668m) compared to other widely used global models such as SRTM (RMSE ±5.666m), NASA (RMSE ±3.656m), and EarthEnv (RMSE ±5.948m), particularly in non-forest and urban environments. However, accuracy can degrade in forested areas, highlighting the importance of considering land cover when choosing a DEM for your design project.

Project Tips

• When choosing a DEM for your design project, research its accuracy and limitations.
• Consider how different land cover types might affect the DEM's performance in your chosen study area.

How to Use in IA

• Reference the study when justifying the choice of a specific Digital Elevation Model (DEM) for your design project, citing its accuracy metrics and suitability for your project's context.
• Use the findings to explain potential limitations or sources of error in your design project if you are using a DEM that is not the most accurate.

Examiner Tips

• Demonstrate an understanding of the limitations of freely available geospatial data and how these limitations can impact design outcomes.
• Justify the selection of specific data sources by referencing relevant research on their accuracy and suitability.

Independent Variable: Type of Digital Elevation Model (MERIT, NASADEM, EarthEnv, SRTM)

Dependent Variable: Vertical accuracy (measured by RMSE against GNSS and reference DEMs)

Controlled Variables: Geographic location (northern Peninsular Malaysia), land cover types (urban, non-forest, forest), reference data resolution

Strengths

• Utilized a large number of GNSS validation points.
• Assessed accuracy across multiple distinct land cover types.

Critical Questions

• To what extent do these findings generalize to other regions with different geological formations and vegetation densities?
• What are the specific error characteristics (e.g., bias, random error) of each DEM, beyond just RMSE?

Extended Essay Application

• An Extended Essay could investigate the impact of DEM accuracy on a specific environmental modelling task, such as flood risk assessment, by comparing results generated using different DEMs.
• Further research could explore methods for improving the accuracy of global DEMs in challenging environments like dense forests.

Source

Vertical Accuracy Assessment of Improvised Global Digital Elevation Models (MERIT, NASADEM, EarthEnv) Using GNSS and Airborne IFSAR DEM · International Journal of Geoinformatics · 2023 · 10.52939/ijg.v19i12.2979