Reference evapotranspiration (ET₀) is an essential component in hydrological and ecological processes. The Penman–Monteith (PM) model of Food and Agriculture Organization of the United Nations (FAO) model requires a number of meteorological parameters; it is urgent to develop high-precision and computationally efficient ET₀ models with fewer parameter inputs. This study proposed the genetic algorithm (GA) to optimize extreme learning machine (ELM), and evaluated the performances of ELM, GA-ELM, and empirical models for estimating daily ET₀ in Southwest China. Daily meteorological data including maximum temperature (T_max), minimum temperature (T_min), wind speed (u₂), relative humidity (RH), net radiation (R_n), and global solar radiation (R_s) during 1992–2016 from meteorological stations were used for model training and testing. The results from the FAO-56 Penman–Monteith formula were used as a control group. The results showed that GA-ELM models (with R² ranging 0.71–0.99, RMSE ranging 0.036–0.77 mm·d⁻¹) outperformed the standalone ELM models (with R² ranging 0.716–0.99, RMSE ranging 0.08–0.77 mm·d⁻¹) during training and testing, both of which were superior to empirical models (with R² ranging 0.36–0.91, RMSE ranging 0.69–2.64 mm·d⁻¹). ET₀ prediction accuracy varies with different input combination models. The machine learning models using T_max, T_min, u₂, RH, and R_n/R_s (GA-ELM5/GA-ELM4 and ELM5/ELM4) obtained the best ET₀ estimates, with R² ranging 0.98–0.99, RMSE ranging 0.03–0.21 mm·d⁻¹, followed by models with T_max, T_min, and R_n/R_s (GA-ELM3/GA-ELM2 and ELM3/ELM2) as inputs. The machine learning models involved with R_n outperformed those with R_s when the quantity of input parameters was the same. Overall, GA-ELM5 (T_max, T_min, u₂, RH and R_n as inputs) outperformed the other models during training and testing, and was thus recommended for daily ET₀ estimation. With the estimation accuracy, computational costs, and availability of input parameters accounted, GA-ELM2 (T_max, T_min, and R_s as inputs) was determined to be the most effective model for estimating daily ET₀ with limited meteorological data in Southwest China.

Investigating the carbon flux in orchard ecosystems is crucial for assessing agroecosystem productivity and optimizing management practices. We measured and estimated carbon fluxes (gross primary productivity, GPP; ecosystem respiration, R_e; and net ecosystem exchange, NEE) and environmental variables in a seven-year-old kiwifruit orchard over two years. Our results showed that diurnal carbon fluxes exhibited bell-shaped patterns, peaking between 12:30 and 15:30. Daily carbon fluxes exhibited a seasonal trend, characterized by an increase followed by a decrease. The average daily GPP, R_e, and NEE values were 6.77, 4.99, and −1.79 g C m⁻² d⁻¹ in 2018, and 5.88, 4.78, and −1.10 g C m⁻² d⁻¹ in 2019, respectively. The orchard sequestered −444.25 g C m⁻² in 2018 and −285.77 g C m⁻² in 2019, which accounted for 26.4% and 18.6% of GPP, respectively. Diurnal GPP and NEE were significantly influenced by photosynthetically active radiation (PAR), with direct path coefficients of 0.75 and 0.88 (p < 0.01), while air temperature (T_a) significantly affected GPP and NEE through PAR, with an indirect path coefficient of 1.12 for both. PAR had a similar effect on daily GPP and NEE, while both were indirectly influenced by soil temperature (T_s) at a 5 cm depth and vapor pressure deficit (VPD). R_e was primarily impacted by VPD, with a direct path coefficient of 0.64 (p < 0.01), while T_a and the concentration of air carbon dioxide (CCO₂) significantly affected GPP through VPD, with indirect path coefficients of 0.82 and −0.80. The leaf area index (LAI) and soil water content (SWC) at a 20 cm depth exhibited a significant correlation with carbon fluxes during the vigorous growing period.