The feeding behavior and fodder intake of cattle, sheep and goats grazing semi-arid sub-Saharan Africa rangelands in Senegal was studied over a one-year cycle. The objective of the study was to improve the assessment of fodder intake by livestock grazing rangelands in semi-arid sub-Saharan Africa. The study was conducted in the service area of a borehole that includes 354 pastoral settlements and a total of 11,000 cattle and 1800 small ruminants. The grazing behavior of five steers was monitored monthly. Their forage intake was estimated every month from hand plucking simulations of all the bites of forage taken by the steers 24 h out of 48. The feces they excreted during the day were collected in fecal bags. Near infrared spectrum (NIRS) images were used to analyze 80 samples of these feces, and the data were then added to an existing reference data set to predict fodder intake and digestibility using NIRS. The complete and contextualized data base was then used to predict the daily fodder intake and its digestibility by cattle, sheep and goats whose fresh feces were systematically and randomly sampled every month in the study area (768 samples). Annual means of the daily dry matter fodder intake were 68.4¿±¿6.4¿g/kg LW0.75 or 17.2¿±¿1.7¿g/kg LW for cattle, 73.1¿±¿5.3¿g/kg LW0.75 or 34.3¿±¿2.8¿g/kg LW for sheep, and 74.7¿±¿8.7¿g/kg LW0.75 or 37.1¿±¿4.4¿g/kg LW for goats. The annual mean organic matter digestibility of grazed fodder was higher for goats (67.4¿±¿3.9%) than for sheep (64.1¿±¿3.2%) and cattle (59.3¿±¿3.8%). The seasonal variations in diet digestibility were bigger for cattle than for sheep, and smaller for goats. The standard reference used to assess livestock fodder intake in tropical Africa relies on a fixed daily intake rate per unit live weight of 25¿g/kg LW. Our results show that this method overestimates (+45%) intake by cattle, especially in the dry season, and underestimates intake by small ruminants (–24%). Alternative assessment methods are p