Fruit weight and yield estimation models for five avocado cultivars in Ethiopia

AbstractAvocado (Persea Americana Mill.) is an important tree crop and proved to be a very rewarding commercial crop for export and local consumptions. Due to high local and global market demand, avocado plantation and production is considerably increasing in many parts of Ethiopia. Thus, pre-harvest yield estimation is necessary for market planning, monitoring, and tracking long-term productivity. Though there are studies on various characteristics of avocado tree, not much has been documented about its productivity due to lack of easy and non-destructive ways to estimate yield. Hence, this study was conducted to develop cultivar-specific and mixed-cultivar fruit weight estimation models using single and multiple predictors including fruit length and diameter. We harvested and weighted 1800 fruits from five cultivars grown in 30 smallholder farmlands. The highest mean fruit load was recorded for Nabal (281 (±36 Fruit. tree−1) and Hass 249 (±27 Fruit. tree−1). A large proportion of avocado fruits were in the range that are commercially valuable. The correlation between fruit load and crown diameter were positive for each cultivar and it was stronger for Ettinger (r = 0.7,p < 0.01, n = 30), followed by Nabal (r = 0.64,P < 0.01, n = 30). We observed that fruit weight correlates significantly (P < 0.01) with fruit diameter and length. Our best models explained 80% for Ettinger, 78% for Fuerte, 80% for Hass, 63% for Nabal, 72% for Reed and 72% for multiple cultivars. Estimated yield was nearly 45 kg. tree−1for Ettinger, Fuerte, Hass, Reed and 90 Kg. tree−1for Nabal. The equations developed in this study are suitable for conducting pre-harvest yield estimation and monitoring changes in avocado productivity. Such information is critical for planned marketing and support government initiatives towards making horticulture sector as a possible engine for poverty alleviation and national economic growth.