The need for accurate estimates of reference crop evapotranspiration (ETo) is important in irrigation planning and design, irrigation scheduling, reservoir management among other applications. ETo can be accurately determined using the internationally accepted FAO Penman–Monteith (FAO-56 PM) equation. However, this requires numerous observed data, including solar radiation, air temperature, relative humidity, and wind speed, which in most cases are unavailable, particularly in developing countries such as the Philippines. This study developed models based on Support Vector Machines (SVMs) and Extreme Learning Machines (ELMs) for the estimation of daily ETo using different input combinations of meteorological data in Region IV-A, Philippines. The performance of machine learning models was compared with the different established alternative empirical models for ETo. The results show that the SVM and ELM models, with at least T_max, T_min, and R_s as inputs, provide the best daily ETo estimates. The accuracy of machine learning models was also found to be superior compared to the empirical models given with same input requirements. In general, SVM and ELM models showed similar modeling performance, although the former showed lower run time than the latter.

Soybean (Glycine max [L.] Merr) is an important crop, as both food for humans and feed for livestock in Cambodia, but the yields are low, due to use of low yielding genotypes and limited use of inputs. This study aimed to investigate the effects of different genotypes and different N and P fertilizer sources on growth, seed yield, and seed protein of soybean. Five genotypes (Sbung, Hongry, 98C81, ACS1, and Kaiabi) were grown under different N (90 kg N ha⁻¹ from urea, nano-n, Nitroplus, and without N application as control) and P fertilizers (50 kg P ha⁻¹ from Inorganic P, nano-p, Mykovam, and without P application as control) in two screenhouse experiments. Shoot, root, and nodulation traits, as well as seed yield and protein, were significantly affected by genotypes and different N and/or P fertilizer sources. Notably, while genotypes Sbung and 98C81 showed the highest yields among the genotypes, regardless of different N and P fertilizers, only Sbung had the highest seed protein. The application of different N and/or P sources significantly increased seed yield, compared to non-N and -P application. Specifically, seed yield and seed protein were higher when treated with nano-n or urea, and number of nodules, root biomass, and nodule dry weight was increased with Nitroplus, whereas seed yield and protein were higher with nano-p, Mykovam, and inorganic P. Application rate of 90 kg N ha⁻¹ and 50 kg P ha⁻¹ produced higher seed yield and its components and seed protein. We conclude that N and P application as well as Nitroplus inoculation can help increase seed yield of soybean in Cambodia.

A study was undertaken in Koun Mom district of Ratanakiri province in Cambodia to analyze the perceptions of the current status and constraints to soybean production and identify solutions to improve production and the management practices. Primary data were collected by personal interviews at field level of 130 producers. Most respondents were in the medium age category, Grade 4 education, with an average land holding of 2.96 ha and annual income of KHR 6,195,548 Riels (about 1548 USD). In terms of economic and production constraints, the high cost of fertilizers, severe insect and disease infestation, were identified as most important. Association with independent characteristics and scientific orientation were not significant, but education, land holding, annual income, socioeconomic status, and risk preference were significantly associated with constraints to soybean production. Within the context of sustainable agricultural production practices, it is suggested to improve high-yielding genotypes, ensure timely availability of high-quality seeds, and identify appropriate crop management practices (planting dates, planting density, nutrient and water management practices) and find ways to efficiently and effectively disseminate information to farmers to enhance soybean production in the region. In addition, extension agents and other agencies should provide soybean farmers marketing information, establish viable links between farmers and relevant stakeholders and private sector to improve access to inputs and modern technologies while the local and state governments should establish rural markets with good market infrastructure to enable farmers have high returns from soybean production.

Bacterial stalk rot (BSR) caused by Dickeya zeae is one of the important diseases of maize that significantly affects maize yield performance. Resistance to D. zeae is influenced by high humidity and temperature. Affected tissues are described as soft, mushy, and emitting a foul odor. Yield losses can reach approximately 98.8% of the grower‟s potential. Quantitative trait locus (QTL) mapping experiments using seven biparental populations were conducted at Syngenta Philippines, Inc., from 2014–2020 to locate consistent QTL and markers involved in BSR resistance. The QTL detected in NMM033, NMM073, NMM089, NMM090, NMM091, NMX003, and NMX001 populations were used to estimate the numbers and positions of consensus QTL with BioMercator V4.2.3 software. Meta-analysis for BSR resistance was conducted by considering all QTL for BSR resistance traits identified in 2014–2020. Among the 49 distinct markers on chromosomes (chrs) 1 to 10, eight most significant loci were detected, i.e., MSRQTL1-1, MSRQTL2-1, MSRQTL3-1, MSRQTL3-2, MSRQTL5-1, MSRQTL5-2, MSRQTL6-1, and MSRQTL10-1. Meta-QTL were identified in chrs 1, 5, and 10 in four populations; in chr 2 in three populations; and in chrs 3 and 6 in two populations evaluated in this study. The regions identified in chrs 1, 2, 3, 5, 6, and 10 with high QTL colocalization across biparental populations were considered as important QTL for BSR resistance traits. Further implementation through fine-mapping is recommended for marker development. The impact of this discovery would strengthen downstream applications in marker-assisted backcrossing and is not only limited to maize BSR resistance but also to other native traits of different crops.