I am currently undertaking a project attempting to explore how morphology along the Holderness Coast, England, has changed over time. I have LiDAR tile data from 2001 - 2017 all at 2m spatial resolution (DTM and DSM) and LiDAR point cloud data from 2008 - 2017. I have tried to use surface creation tools such as 'Slope' and 'Cut-Fill' etc but have not had great results. Ideally I would like to quantify the change in sediment volume over the time but not sure what is the best way to go?
Currently I have created a fishnet that overlays LIDAR data set, then calculated the change height for each zone in the grid but i feel there is a more sophisticated analysis which I could undertake?
Normally, you will perform this analysis using raster data. If the DTM contains enough detail (with the 2m² resolution) you would use this. Do you only have the DTM for 2001 and 2017 or do you also have multiple years? In case of only having the DTM for 2017 and 2001 (say you named those DTM2001 and DTM2017) you would calculate the difference (Minus—Help | ArcGIS Desktop ) DTM2017 - DTM2001. Each pixel in that case will indicate the increment in elevation (perhaps due to sediment as you mentioned). A value of +1 will indicate that in that pixel of 2x2m the volume increased 2*2*1= 4m³, a negative value will indicate a reduction of volume.
In case you have a polygon or a raster that indicates the area of interest, you can use the Zonal Statistics—Help | ArcGIS Desktop tool to get the total volume of change in that zone.
Since this will combine both the reduction and increment in volume, you may want to use a Con—Help | ArcGIS Desktop statement in the Raster Calculator to separate the positive and negative values (of the result of DTM2017 - DTM2001) first before doing the zonal statistics.
I do have multiple dates (nearly every year between 2001 and 2017). For this i used the 'Minus' to calculate the difference for each year i.e. 2001-2002, 2002-2003, 2003-2004 etc. I've created a polygon around my study area so hopefully the zonal statistics should work if applied for each year, thanks!
I expected that using point cloud could be more useful for finding the volume in a 3D environment, do you have any idea?
In case your point cloud has a point density that s high enough to generate and DTM with a higher resolution, you could use the point cloud. The LiDAR data may contain additional information that is not within the DTM or DSM (classification, returns, intensity etc that could help in some processes). However, normally you will need to do some post processing and typically volume changes is easier calculated using the derived rasters (DTM and DSM).