Water quality of Litani River was deteriorated due to rapid population growth and industrial and agricultural activity. Multivariate analysis of spatio-temporal variation of water quality is useful to improve the projects of water quality management and treatment of the river. In this work, analysis of samples from different locations at different seasons was investigated. The spatio-temporal variation of physico-chemical parameters of the water was determined. A total of 11 water quality parameters were monitored over 12 months during 2018 at 3 sites located in different areas of the river. Multivariate statistical techniques were used to study the spatio-temporal evolution of the studied parameters and the correlation between the different factors. Principal Component Analysis (PCA) was applied to the responsible factors for water quality variations during wet and dry periods. The multivariate analysis of variance (MANOVA) was also applied to the same factors and gives the best results for both spatial and temporal analysis. A black point of agricultural, industrial and sewage water pollution was identified in Jeb-Jennine station from the high concentrations of ammonia, sulfate and phosphate. This difference was proved by the major changes in the values of the parameters from one station to the other. Jeb-Jennine represents a main pollution area in the river. The high ammonia, sulfate and phosphate concentrations result from the important agricultural, industrial and sewage water pollution in the area. A high bacterial activity was highlighted in Jeb-Jennine and Quaroun stations because of the presence of the high nitrite concentrations in the two locations. All parameters are highly affected by climate factors, especially temperature and precipitation. TDS, salinity, electrical conductivity and the concentrations of all pollutants increase during wet season affected by the runoff. Other factors can affect the water quality of the river for example geographical features of the region and seasonal human activity like tourism. The correlation between different parameters was evaluated using PCA statistical method. This correlation is not stable, and evolves between wet and dry season.