The objective of this work was to evaluate the productivity and yield of the first four crops of coffee plants (Coffea arabica L.) cultivated under irrigation levels. The experiment was carried out at Fazenda Experimental do Glória, in the County of Uberlândia, MG, Brazil. It was used the randomized blocks design, with eight treatments and four repetitions. The irrigation levels were 0 (without irrigation), 30, 60, 90, 120, 150, 180 and 210% of the evaporation measured in Class A tank (ECA). Planting took place in January 2001, at 3.5x0.7 m spacing. Plots consisted of three rows with eight plants each. A drip irrigation system was used, with self compensating drippers at 3.5 L h-1 flow. Maximum productivity was 115 sacks ha-1, obtained under 164.1% ECA. Replenishment of 143% ECA led to optimum yield 291.8 L of "farm coffee" per processed sack. The effect of irrigation levels on productivity and yield depends on the year. Irrigation does not mitigate the production biannuality of coffee plants, regarding plants cultivated without irrigation.
Coffea arabica; biannuality of production; water deficit; drip irrigation
