Willy William Pazmiño Pachay

Mr. Willy William Pazmiño Pachay

orcid

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
6 November 2024 in Agriculture
Verified
Journal Article
Open access
PDF file
Victor Hugo Morales PeñaVictor Hugo Morales Peña
Victor Hugo Morales Peña
AMArgenis Mora
Argenis Mora
EFElias de Melo Virginio Filho
Elias de Melo Virginio Filho
MVMario Alberto Villatoro-Sánchez
Mario Alberto Villatoro-Sánchez
WPWilly William Pazmiño Pachay
Willy William Pazmiño Pachay
EAEsteban Chanto Ares
Esteban Chanto Ares
This study focused on the phenology of Coffea arabica var. Esperanza L4A5, an F1 interspecific hybrid obtained by crossing commercial varieties with wild genotypes from Ethiopia and Sudan. Most phenological studies on C. arabica have been conducted in traditional high-altitude regions, leaving a gap in the understanding of its behavior in non-traditional areas such as the Caribbean region of Costa Rica. To establish a baseline on the phenological behavior of the Esperanza L4A5 hybrid in this region, we conducted a four-year study examining the effects of different agroforestry associations: (1) Albizia saman; (2) Hymenaea courbaril and Erythrina poeppigiana; (3) Anacardium excelsum and Erythrina poeppigiana; and coffee plots under full sun. Additionally, the phenology of the coffee plants was evaluated under differentiated fertilizations (physical, chemical, and without fertilization), considering meteorological factors such as temperature, humidity, and rainfall. The observed variables included the development of floral nodes, pre-anthesis, anthesis, and fruiting stages. To analyze the relationships between environmental factors, tree cover, fertilization, and the phenological stages, we employed multiple linear regression (MLR), which revealed that both tree cover and physical and chemical fertilizations had significant effects on the presence of developed floral nodes and, consequently, on fruit production. Furthermore, the random forest (RF) model was applied to capture complex interactions between variables and to rank the importance of meteorological factors, tree cover, and fertilization practices. These analyses demonstrated that the Esperanza L4A5 hybrid exhibited viable phenological development under the atypical conditions of the Caribbean region of Costa Rica, suggesting its potential to adapt and thrive in non-traditional coffee-growing areas.

8 October 2026 in Preprints.org
Verified
Preprint
Open access
PDF file
Victor Hugo Morales PeñaVictor Hugo Morales Peña
Victor Hugo Morales Peña
WPWilly William Pazmiño Pachay
Willy William Pazmiño Pachay
Oswaldo Vicente Navarrete CarreñoOswaldo Vicente Navarrete Carreño
Oswaldo Vicente Navarrete Carreño
CCCarly CHERY
Carly CHERY
AMArgenis Mora
Argenis Mora
Accurate yield prediction in perennial crops is challenging when hierarchical datasets contain repeated response values because lower-level evaluation can inflate apparent model performance. We re-audited a four-year Coffea arabica var. Esperanza L4A5 dataset from a lowland agroforestry trial in Costa Rica. The 2020–2022 development dataset comprised 45 response-unit-year records representing 15 recurring management × fertilization combinations, while the corresponding 15 combinations in 2023 were retained as a temporal holdout. Decision Tree, Random Forest (RF), Support Vector Regression (SVR), and ridge regression were evaluated using forward temporal validation and compared with training-mean and previous-year persistence benchmarks. On the 2023 holdout, SVR (R2 = 0.53; RMSE = 0.308 kg plant−1) and RF (R2 = 0.51; RMSE = 0.312 kg plant−1) improved on the training-mean baseline, but previous-year yield persistence performed substantially better (R2 = 0.96; RMSE = 0.091 kg plant−1). Leakage-aware aggregation therefore materially changed conclusions about model performance, and added algorithmic complexity did not improve prediction beyond recent yield history in this short time series. Longer multi-site time series and genuinely pre-harvest predictors are required before operational forecasting can be justified.