Your question is somewhat vague: It depends on what you mean by "best". Cheapest? Most accurate? Fastest turnaround time? Least amount of sample required? Able to be performed with the equipment you have available?

Soil samples are harder to digest completely than plant samples, so there is no single "best" method. It will depend greatly on the type of soil (sand% vs loam% vs clay% for a start). Plant samples: grains/seeds/leaves/roots/wood/bark?

I'm not aware of exactly how Uranium responds in nitric acid digests, but I believe it can be "sticky" in the PTFE sample tubing of the ICPMS, so may cause memory issues and carry over from high concentration samples to low concentration.

A quick GoogleScholar search showed up the following papers regarding U and soil/plants:

Bellis et al 2001 https://doi.org/10.1016/S0269-7491(00)00236-0

Oliver et al 2007 DOI 10.1039/B700719A

Rufyikiri et al https://doi.org/10.1016/j.envpol.2005.08.072

I suggest you read these and a few other papers and see what sort of materials you will need and whether or not it can be replicated at your laboratory. The "best" method is no use if you can't replicate it at your facility.

@Matthew Wheal

I mean the most accurate in order to measure the total concentration of this element in the sample

Thanks alot for your answers