Spatially Distributed Hydrological Modelling: Wetness Derived from Digital Elevation Models to Estimate Peatland CarbonHasan, Abdulghani
To study the hydrology of peatlands and explore wetness distribution is difficult
mainly due to the complexity of the surface of peatlands, and also due to the presence of permafrost underlain peatlands in the arctic regions. I have chosen the area called Stordalen mire in the arctic region in northern Sweden for my study.
In this thesis, I aimed to study spatially distributed hydrological modelling in general, focusing mainly on evaluation and developing tools that can be used to improve wetness estimation using Digital Elevation Models (DEMs). The estimated wetness can be used as an input for peatland carbon models.
DEMs with different resolutions are created using high resolution LiDAR data.
Different search radiuses are used in the interpolations. The accuracy of the generated DEMs is studied to select the most accurate DEM for each selected resolution. The search radius, but not the cell size, significantly influences the accuracy of a DEM, and the accuracy is generally higher the shorter the interpolation search radius. DEM resolution versus topographic wetness index variables (i.e. slope and drainage area) is studied. Slope values become lower and drainage area values become higher when the resolution decreases. Further, a study of DEM accuracy related to different slopes is also carried out and shows that the errors in elevation are greater when the terrain is steep than when it is flat.
A new triangular form-based multiple flow distribution and flow accumulation
algorithm (TFM) was created in this study. We have estimated flow distribution by using our new TFM algorithm. With this TFM algorithm, it becomes possible to deal with artefacts that normally interrupt flow distribution, like flat areas, sinks and manmade structures. This will help to overcome the complexity of peatland hydrology.
The results of comparing our new algorithm with other well-known algorithms used in most GIS programs show that the TFM algorithm produces more realistic results than other algorithms. Testing shows the capability of the new TFM algorithm to distribute the flow in different terrain types, flat areas and sinks, and makes it suitable for simulating real flow distribution over any surface/terrain.
Topographic wetness index (TWI) was estimated for the study area using our new
flow distribution and flow accumulation algorithm TFM. Estimating TWI values
depending only on DEMs is a very cost-effective method that can be used to estimate wetness data required for the modelling of peatlands. A permafrost model was created to demonstrate the possibility of using an analytically based approach with semiempirical equations to estimate the maximum thawing depth (active layer thickness) above permafrost. Field work using water level sensors was carried out to measure the temporal fluctuation of water surface. The field work water level measurements led to better understanding of flow regime in the peatlands, especially when a seasonally frozen layer or permafrost lies under it. The field work also helped to confirm that estimated wetness using the proposed flow routing algorithm on digital elevation model can be used to distribute wetness to all cells in a DEM.
Keywordsmodelling hydrology; digital elevation models; topographic wetness index; LiDAR data; peatland; permafrost; Stordalen
Publisher: Lund University
UKÄ Subject classification
URI (permanent link to this page)