Genotype by environment cultivar evaluation for cassava brown streak disease resistance in Tanzania

Cassava brown steak disease (CBSD), caused by Cassava brown streak virus (CBSV) and Ugandan cassava brown streak virus (UCBSV), is the most important biotic constraint to cassava production in East and Central Africa. Concerted efforts are required to prevent further spread into West Africa as well as to reduce losses in areas already affected. The study reported here was part of a five-country (Kenya, Malawi, Mozambique, Tanzania and Uganda) programme that aimed to identify superior cassava cultivars resistant to CBSD and to disseminate them widely in the region. Seventeen tissue-cultured and virus-tested cultivars were evaluated in Tanzania across nine sites with diverse CBSD inoculum conditions. Experiments were planted using an alpha-lattice design and assessments were made of surrounding inoculum pressure, CBSD foliar and root incidence and root yield at harvest. There were large differences in CBSD infection between sites, with greatest spread recorded from the north-western Lake (Victoria) zone. Differences were driven by Bemisia tabaci whitefly vector abundance and CBSD inoculum pressure. Both CBSV and UCBSV were almost equally represented in cassava fields surrounding experimental plots, although CBSV predominated in the north-west whilst UCBSV was more frequent in coastal and southern sites. However, the incidence of CBSV was much greater than that of UCBSV in initially virus-free experimental plots, suggesting that CBSV is more virulent. Cultivars could be categorised into three groups based on the degree of CBSD symptom expression in shoots and roots. The seven cultivars (F10_30R2, Eyope, Mkumba, Mkuranga1, Narocass1, Nase3 and Orera) in the most resistant category each had shoot and root incidences of less than 20%. Fresh root yield differed between sites and cultivars, but there was no genotype by environment interaction for this trait, probably attributable to the large fertility and soil moisture differences between sites. Susceptible cultivars and the local check performed well in the absence of CBSD pressure, highlighting the importance of exploiting quality and yield traits of local landraces in breeding programmes. Overall, our results emphasized the importance of applying a balanced strategy for CBSD management. This should use both improved and local germplasm resources to generate high yielding cultivars for specific end-user traits, and combine the deployment of improved cultivars with phytosanitary control measures including the use of healthy planting material and planting during periods of reduced CBSD infection.