With regard to the vicinity of structures a number of approaches may be taken. Some sort of monitoring is often used to make predictions, typically in the form of repeat bathymetry surveys and surface difference analysis. Secondly hydrodynamic simulations capable of calculating sediment transport and scour can be utilised to predict the rate and magnitude of bathymetric change at structures.
You can also use altimeters or similar instruments mounted to the structures which detect local changes in bed elevation, which might be cheaper. For scour analysis I suggest you read, "Scour Technology: Mechanics and Engineering Practice (McGraw-Hill Civil Engineering) by George W. Annandale (Nov 15, 2005)"
difficult to say. it depends on what you mean for long term. try to estimate the average long shore sediment transport and rate of accumulation. if you will be able to correlate it with several shoreline position through time you may find the possible relation for predictions. it also depend on what are the forces influencing long shore drift and if within the sand sharing system you can assume costant some of antropic action (dredging; management of dams; construction of other structures that may change the long shore sediment drift). good luck
Should you look at the bathymetric changes in a well-defined, yet sizeable, area (vicinity of a structure) over time, I would agree with the different methods put forward by the people above, but I would personally start by analysing the bathymetric profiles recorded over the years, as well as any records of any mapping sensor (e.g. Side Scan Sonar or satellite images should the water be clear enough and the bottom reflective enough). I would try to correlate them, map changes in shape to known patterns of sediment accumulation and/or erosion. Thus I would try to calibrate results with on the spot field measurements using well located sensors to detect local changes in sediment beddings, elevation, over an appropriate period of time. This time period could also be defined by analyzing the records cited above through time if records exist. I would also set up sediment traps at regular intervals (horizontally and vertically) to calibrate the sediment type and granulometry. Good luck.
My PhD Thesis is about the development of analytical, statistical and numerical models which assess beach evolution in the vicinity of coastal constructions. You could have a look at my published papers and the included references.