Because the shell and/or surrounding of nanoparticle influence LSRP and DSL differently. As said above, DLS measures the hydrodinamic size of nanoparticle; that is physical size of nanoparticle with its shell as it would be measured by caliper. UV-VIS gives the size of nanoparticle considering it as a core-shell with defined electronic/optical properties and characterises the size of plasmonic nanoparticle in media with some effective refractive index. Therefore, two nanoparticles of the same physical size and diffusion coefficient may have different optical spectra (and vice versa).

Probably because your are seeing different sizes, in the DLS you can observe the hydrodinamic size of the particle, while in the UV-Vis you get a value that would be the "real" size, probably you should use a microscopy technique to confirm which one is correct.

Journal of Nanomaterials

Volume 2013 (2013), Article ID 313081, 10 pages

Detection Limits of DLS and UV-Vis Spectroscopy in Characterization of Polydisperse Nanoparticles Colloids

Colloids characterized by atomic force microscopy (AFM), transmission electron microscopy (TEM), DLS, and UV-Vis spectroscopy

Because the shell and/or surrounding of nanoparticle influence LSRP and DSL differently. As said above, DLS measures the hydrodinamic size of nanoparticle; that is physical size of nanoparticle with its shell as it would be measured by caliper. UV-VIS gives the size of nanoparticle considering it as a core-shell with defined electronic/optical properties and characterises the size of plasmonic nanoparticle in media with some effective refractive index. Therefore, two nanoparticles of the same physical size and diffusion coefficient may have different optical spectra (and vice versa).

Hydrodynamic particle size from DLS by "scattering" and particle size determination from UV-VIS via "scattering and absorption" will be different.

Polydispersity and anisotropy of the samples can distort the results, so we could not see the real population of particles because big particles can screen smaller ones.

Basically,  DLS measures the hydrodynamic diameter of nanoparticle/nanocluster and it consider the size of solvent cage surrounding the particle in calculating it. So, if you compare the size calculated using TEM  and DLS, DLS data will be  little bit more. So far UV-vis is concerned, size depends on various factor such as RI, dielectric constant of the medium, presence of capping agent etc. So, measured size will be differ a bit if measured by both techniques.

Yes, the hydration shell may be one difference, another (and probably more significant) is that DLS will by default provide an intensity weighted result, whereas UV-Vis is a mass (or volume) weighted result. Typically the intensity weighted result will be larger than the volume based due to the higher scatter of larger particles within the distribution.

You may try to look the volume distribution result with the UVVis for a more direct comparison, provided the DLS data are of suitable quality.

http://www.materials-talks.com/blog/2014/01/23/intensity-volume-number-which-size-is-correct/

There is nothing like accurate size at least for colloidal particles. Every method measures something different and assumptions have to be made more or less valid.