Regarding the definition of "surface reflectance" and "remote sensing reflectance": 

- Surface reflectance is a property of the surface (or ground) itself, i.e. in this case the sea surface. It is determined using Fresnel's Equations and Snell's Law, both functions of the relative refractive index (of air and water), which depend on water salinity, temperature, and wavelength of light. The surface reflectance is a function of surface wind-dependent roughness and light incident or radiance distribution respectively. The term tells how much of the light is reflected at the sea surface and into what direction. 

- Remote sensing reflectance (Rrs) contains the spectral colour information of the water body (below the sea surface). Rrs is the ratio between water-leaving radiance (Lw, above the sea surface) and downwelling irradiance (Ed, above the sea surface). Lw can be estimated from above-water radiometric measurements, in this case reflected skylight must be removed using a "surface reflectance factor" (rho). 

Further comment: In ocean colour, often the term "water-leaving reflectance" (rho_w) is used instead of Rrs. The difference is:  rho_w = pi*Rrs (with related change in units). 

http://www.oceanopticsbook.info/view/overview_of_optical_oceanography/reflectances

It is not clear what reason the unpublished study you mention has for suggesting that 6SV cannot be used over water. In a strict sense that statement is obviously untrue (i.e. you can process an image containing water surfaces with a 6SV-derived algorithm), so perhaps it instead suggests that it should not be done?

With respect to remote sensing reflectance vs. surface reflectance, the document you attached has a more detailed explanation than any you will get on researchgate... and it also explains why you may be able to convert one to the other (in simplest term divide/multiply by pi), but will be unable to do so correctly unless you have information on the angular distribution of the upwelling light field. This may be the reason for the statement in the unpublished study - 6SV assumes Lambertian ground surfaces, and water does not fit that assumption well.

As mentioned above there are differences when correcting over water versus land and so you can use a land focused AC over water, but the assumptions made will be causing errors. Also, you'll not be correcting for effects such a sky and sun glint plus adjacency. Over water these errors can become a significant issue as the reflectance of water is much lower, so AC code specifically designed for above-water applications tends to be used.

The pi conversion is very much a simplification, with a f/Q factor (described in the document) traditionally used in marine AC code and Park and Ruddick (2005) improving on this further for coastal waters.

Regarding the definition of "surface reflectance" and "remote sensing reflectance": 

- Surface reflectance is a property of the surface (or ground) itself, i.e. in this case the sea surface. It is determined using Fresnel's Equations and Snell's Law, both functions of the relative refractive index (of air and water), which depend on water salinity, temperature, and wavelength of light. The surface reflectance is a function of surface wind-dependent roughness and light incident or radiance distribution respectively. The term tells how much of the light is reflected at the sea surface and into what direction. 

- Remote sensing reflectance (Rrs) contains the spectral colour information of the water body (below the sea surface). Rrs is the ratio between water-leaving radiance (Lw, above the sea surface) and downwelling irradiance (Ed, above the sea surface). Lw can be estimated from above-water radiometric measurements, in this case reflected skylight must be removed using a "surface reflectance factor" (rho). 

Further comment: In ocean colour, often the term "water-leaving reflectance" (rho_w) is used instead of Rrs. The difference is:  rho_w = pi*Rrs (with related change in units). 

http://www.oceanopticsbook.info/view/overview_of_optical_oceanography/reflectances

Dear All,

Many thanks for your valuable comments. These helped me to understand the things more clearly.

Regards

Dear Majid, 

I think that you can use FLASSH from ENVI that is a good alternative to 6SV model, like says in the article that I attach results are similars. Also you should consider that it is better to use DOS if you have a the most of image with water.

Conference Paper Comparison between empirical and physically based models of ...

https://www.researchgate.net/publication/301797360_Analyzing_land_surface_temperature_distribution_in_response_to_land_useland_cover_change_using_split_window_algorithm_and_spectral_radiance_model_in_Sundarban_Biosphere_Reserve_India

Article Analyzing land surface temperature distribution in response ...

The surface reflectance mean the energy received and recorded by sensor platform on satellite from surface of the body whereas remote sensing reflantance mean the scattering of the energy path due to climatic condition i.e cloudy, moisture in atmospheric, dust particle, time of recording, angle of incidence while recording etc received toward the sensor. These are both terminology are difference. The reflanrtance received from object is could not rectified while atmospheric correction can be rectified by applying different methods.

Surface reflectance and remote sensing reflectance should be the same. You can use surface reflectance to correct remote sensing reflectance as ground truth data. There is always difference due light wave travailing through the atmospheric layers and due to distance. 

regards

Maybe, the use of the "reflectance factor" term, as in many papers, be convenient to the describe your data and can be compared another spectro in the literature.

Following

The main difference lies in the basic definition of both properties,

Surface reflectance is the property of opaque surface or in other words you can say diffused radiation for that surface which is hemispherical in nature but remote sensing reflection is the part of that diffused radiation in the direction of solid angle of the observer or in this case sensor so if you have remote sensing reflection than we simply need to integrate that for hemispherical surface which finally gives us π (pi) factor after cancelling all terms.

Finally

Surface reflectance = π * remote sensing reflection.

السلام عليكم. وبخصوص السؤال فان ال Surface reflectance هو مايسجله المتحسس مما يعكسه سطح الهدف فعليا فقط. اما ال Remote sensing reflectance فهو الانعكاس المذكور اعﻻه اضافة الى اي انعكاسات ناشئة في الفضاء بين الهدف والمتحسس ،،، ودمتم