r.mblend.html

<h2>DESCRIPTION</h2>

This module implements Mblend, a DEM merging method proposed by Leit&atilde;o <i>et al.</i> (2016). It deals with cases where a study area is only partially covered by a high resolution DEM, with a coarser DEM available for the remainder (as in the case shown below). <i>r.mblend</i> merges the two DEMs, producing a smooth transition from the high resolution DEM to the low resolution DEM.<br><br>

<center>
<img width=40% align=center src=both_inputs.png><br><br>
</center>

The module works by identifying the edge between the two rasters (near edge, shown in read below) and the edge composed by the cells in the low resolution DEM farther away from the high resolution raster (far edge, shown in blue below). To each point along the near edge is assigned the difference between the two DEMs. To each point in the far edge is assigned the value 0. The Inverse Distance Weighted (IDW) method is then used to interpolate a new raster with the points along the two edges. This differences raster thus trends from the full difference at the near edge towards zero at the far edge. <br><br>

<center>
<img width=40% align=center src=edges.png><br><br>
</center>

The differences raster is finally added to the low resolution DEM given as input. In the resulting DEM, cells along the near edge take the values in the high resolution raster. The farther away from the near edge (and closer to to the far edge) the closer is their value is to the low resolution raster, producing a smooth transition, without artefacts.<br><br>

<center>
<img width=40% src=blended.png>
</center>  

<h2>EXAMPLES</h2>

Merge the <tt>best_dem</tt> and <tt>other_dem</tt> raster maps from the current mapset:
<div class="code"><pre>
r.mblend high=best_dem low=other_dem output=result
</pre></div>

Modifying the far edge distance cut-off:

<div class="code"><pre>
r.mblend high=best_dem low=other_dem output=result far_edge=90
</pre></div>
 

<h2>REFERENCES</h2>

J.P. Leit&atilde;o, L.M. de Sousa, <a href=https://doi.org/10.1016/j.jhydrol.2018.04.043>Towards the optimal fusion of high-resolution Digital Elevation Models for detailed urban flood assessment</a>, <i>Journal of Hydrology</i>, Volume 561, June 2018, Pages 651-661, DOI: <a href=https://doi.org/10.1016/j.jhydrol.2018.04.043>10.1016/j.jhydrol.2018.04.043</a>. 
<br><br>
L.M. de Sousa, J.P. Leit&atilde;o, <a href=http://www.scitepress.org/PublicationsDetail.aspx?ID=mcJr0zto14w=&t=1>Improvements to DEM Merging with r.mblend</a>. In <i>Proceedings of the 4th International Conference on Geographical Information Systems Theory, Applications and Management - Volume 1: GISTAM</i>, March 2018, pages 42-49. ISBN 978-989-758-294-3 DOI: <a href=http://www.scitepress.org/DigitalLibrary/Link.aspx?doi=10.5220/0006672500420049>10.5220/0006672500420049</a>. 
<br><br>
J.P. Leit&atilde;o, D. Prodanovic, C. Maksimovic, <a href=http://www.sciencedirect.com/science/article/pii/S0098300416300012>Improving merge methods for grid-based digital elevation models</a>, <i>Computers & Geosciences</i>, Volume 88, March 2016, Pages 115-131, ISSN 0098-3004, DOI: <a href=http://doi.org/10.1016/j.cageo.2016.01.001>10.1016/j.cageo.2016.01.001</a>.


<h2>SEE ALSO</h2>

To report bugs, propose new features or discuss the funcioning of this add-on, visit the project repository at <a href=https://github.com/ldesousa/r.mblend>GitHub</a>.

<h2>AUTHORS</h2>

Lu&#237;s Moreira de Sousa<br>
ISRIC - World Soil Information
<br><br>
Jo&atilde;o Paulo Leit&atilde;o<br>
EAWAG: Swiss Federal Institute of Aquatic Science and Technology.

<p>
<i>Last changed: $Date$</i>