Dear Milad,

although appearing trivial, the differentiation between land and water can be tough, especially if you are dealing with changing coastal environments (e.g. steep coast, flat coast, lagoon). Hence, there is no "most accurate" index, as all available, and I will provide some papers to it, are applicable, but eventually you will come to the point to define a threshold to differentiate the two classes (land and water). This is the decisive act during the processing and whatever kind of water body you intend to extract a global threshold application (one threshold for the entire image) might not work if you have the aforementioned changing conditions. E.g. assume you have steep coasts and next to it a river mouth with a high suspension load, the reflectance values of the water area at the mouth and the deep water will differ, so will the index values, leaving you with the need of two values instead of one to properly detect the coastline at both spots properly.

To provide you with some indices and aiming specifically at Landsat (mentioned n your keywords):

NDWI - McFeeters, S. The use of the Normalized Difference Water Index (NDWI) in the delineation of open water features. Int. J. Remote Sens. 1996, 17, 1425–1432.

MNDWI -  Xu, H. Modification of normalised difference water index (NDWI) to enhance open water features in remotely sensed imagery. Int. J. Remote Sens. 2006, 27, 3025–3033.

AWEI - Feyisa, G.L.; Meilby, H.; Fensholt, R.; Proud, S.R. Automated Water Extraction Index: A new technique for surface water mapping using Landsat imagery. Remote Sens. Environ. 2014, 140, 23–35. 

 WRI - Shen, L.; Li, C. Water Body Extraction from Landsat ETM+ Imagery Using Adaboost Algorithm. In Proceedings of 18th International Conference on Geoinformatics, 18–20 June 2010, Beijing, China; pp. 1–4.

If you need more help, let me know.

All the  best,

Ulf

Will this paper answer your question?

http://web.pdx.edu/~nauna/resources/19-water2000_dec_1461-1467.pdf

Try attached file-sory it is in turkish

Dear Milad,

although appearing trivial, the differentiation between land and water can be tough, especially if you are dealing with changing coastal environments (e.g. steep coast, flat coast, lagoon). Hence, there is no "most accurate" index, as all available, and I will provide some papers to it, are applicable, but eventually you will come to the point to define a threshold to differentiate the two classes (land and water). This is the decisive act during the processing and whatever kind of water body you intend to extract a global threshold application (one threshold for the entire image) might not work if you have the aforementioned changing conditions. E.g. assume you have steep coasts and next to it a river mouth with a high suspension load, the reflectance values of the water area at the mouth and the deep water will differ, so will the index values, leaving you with the need of two values instead of one to properly detect the coastline at both spots properly.

To provide you with some indices and aiming specifically at Landsat (mentioned n your keywords):

NDWI - McFeeters, S. The use of the Normalized Difference Water Index (NDWI) in the delineation of open water features. Int. J. Remote Sens. 1996, 17, 1425–1432.

MNDWI -  Xu, H. Modification of normalised difference water index (NDWI) to enhance open water features in remotely sensed imagery. Int. J. Remote Sens. 2006, 27, 3025–3033.

AWEI - Feyisa, G.L.; Meilby, H.; Fensholt, R.; Proud, S.R. Automated Water Extraction Index: A new technique for surface water mapping using Landsat imagery. Remote Sens. Environ. 2014, 140, 23–35. 

 WRI - Shen, L.; Li, C. Water Body Extraction from Landsat ETM+ Imagery Using Adaboost Algorithm. In Proceedings of 18th International Conference on Geoinformatics, 18–20 June 2010, Beijing, China; pp. 1–4.

If you need more help, let me know.

All the  best,

Ulf

You can use index composition and HIS (hue, intensity and saturation) transformation.

Run the differences indices like

NDVI: Normalised Difference Vegetation Index use to enhance the vegetation feature

MNDWI: Modified normalised difference water index use to enhance the open water feature.

NDBI:the Normalized Difference Built-up Index use to enhance bare land feature or build up area.

Then HIS transformation is apply to extract water bodies clearly demarked boundary and remove hill shadows.

Try this index: Floating Algae Index (FAI).

It was first developed for the MODIS images, and some studies showed that FAI was less sensitive to the environmental variables and more stable to establish a region-wide and potentially time-independent threshold. I am not sure if it can be used for the Landsat. Anyway you can try it.

Hu C. A novel ocean color index to detect floating algae in the global oceans[J]. Remote Sensing of Environment, 2009, 113(10): 2118-2129.

Feng L, Hu C, Chen X, et al. Assessment of inundation changes of Poyang Lake using MODIS observations between 2000 and 2010[J]. Remote Sensing of Environment, 2012, 121: 80-92.

the normalized difference of swir band with red band will probably yield the result for water index like NDVI.

IR lidar does not reflect over water and this can lead to very accurate boundaries

Same for any IR Remote sensing Instrument at high resolution (Say 60cm or 1m Ikonos or similar systems).