Good day,
I am doing some research in river habit modelling and I am looking to see the effect of resolution of topography description has on 2D modelling of river habitat.
If there is anyone who has knowledge in this field of modelling rivers and willing to share their information, please don't hesitate to contact me
Any help would be appreciated
Could you be more specific? What do you mean with "effect on 2D modelling of river habitat"? What do you mean with resolution of topography; cell size? or do you mean the precision with wich the topography is measured?
It is usual practice in any 2D modelling to try out at least two cell sizes to see if it has some effect on the results; that is water level and velocity. But usually more important than the cell size (which of course has an effect on the detail of information you will get), it is the structure of the cell array that has an effect on the previous.
In any case, since a bigger cell size means at the end an averaging of conditions within that region, then I would expect that "small habitat niches" will not be identified when using larger cells.
With more information maybe I can help you some more.
Good day Rafael,
Thanks a lot for getting back to me.
So I am currently doing some research in river hydraulics. The research is being conducted on a river within Johannesburg, South Africa and LIDAR survey has been done for a certain part.
I know that LIDAR is adequate for flood modelling but when it comes to habitat modelling you need more detail so this is where my question comes into play. What are the effects of modelling river habitats using survey data such as:
1) LIDAR
2 If I just go to site and measure X, Y and Z for some main features in the river, say a Huge Bolder, Reed beds, rocks and so on and guess manning's n for the river bed. I would then model the features and the river bed in a custom grid.
The idea I have is to see if it is possible to have a simple survey done on a river (I am taking about low flow depth rivers) as mentioned in option 2 and see if it has any effect on the results, at the same time saving money and time in detailed survey.
There has been lots of research done on modelling river habits but I know how to model the river but I just want to know how much can we reduce the survey until the results are not adequate
I appreciate the help
First:
LIDAR: in principle Lidar does not has to be "low resolution". Just as conventional surveying, you can achieve very detailed results. It is just a matter of how you carry out your survey. But I guess you are talking about some sort of readily available Lidar survey, wich of course has a given resolution that you cannot modify.
Second:
I would start with a differente question: What is or could be a characteristic size of the habitats I am planning to modell? Lets say you are working at some mountain stream with about 2 meters width. A pool in this river system (within a pool / riffle system) is likely going then to be smaller than the whole width of the river. Lets say 1 meter. If you want this niche to be diferentiated from the sorroundings, then your survey hast to have at least this resolution. If you want to know something about the dynamics within this niche, then you need a finer resolution.
Another issue that you may want to take into account is that detail among cross sections (downstream direction) tends to be more important than the detail within one cross sections. The reason: that is the direction of the movement.
Finally, to carry out any hidrological survey efficiently you may want to focus on the break points instead of having a regular network of points.
I would also suggest sampling using high-resolution imagery or an actually visit to a couple of different locations on the river in order to compare visual assessment with the LiDAR features in order to ground-truth the data. This will give you a better idea of feasibility using the available LiDAR data.
Also important to keep in mind is that the reason LiDAR by itself often does not work well for habitat modeling is because the water surface returns will not accurately represent the low flow channel which is crucial for determining available habitat at low flows. Often what is needed is to augment LiDAR with bathymetric surveys of the low flow channel region.
That being said I think that it is an interesting and relevant question to ask: how much low flow topographic information do you need in order to adequately represent habitat. In that respect I think Raphael is right on in asking the question you need to ask before any modeling exercise: What am I trying to model? What Species? What flows? etc.
- Badges
- Science topic
- Similar topics
- Engineering
- Civil Engineering