Figueiredo, Ricardo T. [UNESP]Fernandes, Adalton M. [UNESP]Garreto, Francisca G. S. [UNESP]Silva, Jéssica A. [UNESP]Nunes, Jason G. S. [UNESP]Vargas, Pablo F. [UNESP]2023-07-292023-07-292023-01-01Agronomy Journal.1435-06450002-1962http://hdl.handle.net/11449/248957The response of sweetpotato [Ipomoea batatas (L.) Lam.] to potassium (K) can vary based on soil K availability and the rate and timing of K application. This study evaluated the effects of K application rate and timing on leaf K concentration and yield attributes of sweetpotatoes grown in tropical sandy soils with different K availabilities. Treatments had three rates (60, 120, and 240 kg ha−1 K2O) and four timings (pl, the full rate at planting; pl-1.3, 1/2 rate at planting plus 1/2 at 1.3 months after planting [MP]; pl-3, 1/2 rate at planting plus 1/2 at 3 MP; pl-1.3–3, 1/3 rate at planting plus 1/3 each at 1.3 and 3 MP), including a control (without K application). The benefits of K timings and application rates on sweetpotato yield were greater in K-deficient soils with the maximum yield of sweetpotato (25.8–34.3 Mg ha−1) at the estimated rate of 153–179 kg ha−1 K2O in three application times. In soil with medium exchangeable K, the maximum total root yield (28.5–34.2 Mg ha−1) occurred at an optimum estimated rate of 113–122 kg ha−1 K2O. In soils with high exchangeable K, K application rates reduced the yield of roots or starch, and no K deficiency occurred in the unfertilized K treatments. To obtain high storage root and starch yields in sweetpotatoes grown in K-deficient tropical sandy soils, there should be three K applications, and K application rates should be based on K levels already available in the soil.engArtigo10.1002/agj2.213642-s2.0-85161421183