First, I apologize for putting forward yet another query on the calculation of the "Slope, Aspect, Curvature" module under the Terrain Analysis > Morphometry menu. I wish to confirm on the following aspects:
(a) if the units of plan and profile curvatures, as well as cross-sectional and logitudinal curvatures are still provided in 1/map units?
(Ref: https://sourceforge.net/p/saga-gis/discussion/790705/thread/30c1dfe8/#275f)
(b) What is the dissimilarity (or similarity) between calculation of the plan curvature and that of cross-sectional curvature, and between profile curvature and longitudinal curvature, as I have got two different sets of results (if they are the same as per Zevenbergen & Thorne, 1987)?
(Ref: https://sourceforge.net/p/saga-gis/discussion/790705/thread/30c1dfe8/#275f).
There was a reported change in the curvature calculation options in SAGA GIS revision 1778
( Ref: https://sourceforge.net/p/saga-gis/code-0/1778/)
I wish to simply know about the status of these naming conventions and the units that are currently followed in the most recent version of SAGA GIS.
Best Regards,
Parag
Last edit: Parag Jyoti Dutta 2018-10-05
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Thanks a lot for providing the information on curvatures. You have always been kind and prompt to provide answers to the point.
Here, I would like to provide a useful reference for SAGA users regarding calculation of curvatures when their frequencies, particularly the kurtosis, deviate too much from a 'normal' distribution. Csillik et al. (2015) have provided a normalization formula with supporting results to transform curvatures towards distributions close to the standard 'normal' .
The transformation formula is Transform = arctan (k × curvature)
where k is a parameter to give normalized curvature with kurtosis close to 0. Long tails are pulled in symmetrically, depending on the value of k: the higher the k value, the more pulling in. Typical values for k range from 0.1 to 7.5. The transform preserves the sign (+/-).
Csillik, O., Evans, I.S., Dragut ¸, L., Transformation (normalization) of slope gradient and surface curvatures, automated for statistical analyses from DEMs, Geomorphology (2015), doi: 10.1016/j.geomorph.2014.12.038
Regards,
Parag
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Dear SAGA Users,
First, I apologize for putting forward yet another query on the calculation of the "Slope, Aspect, Curvature" module under the Terrain Analysis > Morphometry menu. I wish to confirm on the following aspects:
(a) if the units of plan and profile curvatures, as well as cross-sectional and logitudinal curvatures are still provided in 1/map units?
(Ref: https://sourceforge.net/p/saga-gis/discussion/790705/thread/30c1dfe8/#275f)
(b) What is the dissimilarity (or similarity) between calculation of the plan curvature and that of cross-sectional curvature, and between profile curvature and longitudinal curvature, as I have got two different sets of results (if they are the same as per Zevenbergen & Thorne, 1987)?
(Ref: https://sourceforge.net/p/saga-gis/discussion/790705/thread/30c1dfe8/#275f).
There was a reported change in the curvature calculation options in SAGA GIS revision 1778
( Ref: https://sourceforge.net/p/saga-gis/code-0/1778/)
I wish to simply know about the status of these naming conventions and the units that are currently followed in the most recent version of SAGA GIS.
Best Regards,
Parag
Last edit: Parag Jyoti Dutta 2018-10-05
Hi Parag,
regarding (a), yes, the units are still 1/map units.
Regarding (b) - please have a look at the cited literature, it shows all the equations which are used in SAGA to derive these curvature parameters:
Olaya, V. (2006): Basic Land-Surface Parameters. In: Hengl, T., Reuter, H.I. [Eds.]: Geomorphometry: Concepts, Software, Applications. Developments in Soil Science, Elsevier, Vol.33, 141-169.
Best regards,
Volker
Hi Volker,
Thanks a lot for providing the information on curvatures. You have always been kind and prompt to provide answers to the point.
Here, I would like to provide a useful reference for SAGA users regarding calculation of curvatures when their frequencies, particularly the kurtosis, deviate too much from a 'normal' distribution. Csillik et al. (2015) have provided a normalization formula with supporting results to transform curvatures towards distributions close to the standard 'normal' .
The transformation formula is Transform = arctan (k × curvature)
where k is a parameter to give normalized curvature with kurtosis close to 0. Long tails are pulled in symmetrically, depending on the value of k: the higher the k value, the more pulling in. Typical values for k range from 0.1 to 7.5. The transform preserves the sign (+/-).
Csillik, O., Evans, I.S., Dragut ¸, L., Transformation (normalization) of slope gradient and surface curvatures, automated for statistical analyses from DEMs, Geomorphology (2015), doi: 10.1016/j.geomorph.2014.12.038
Regards,
Parag