Present climate and socioeconomic issues threaten global food and environmental security. The weakening situation of domestic agriculture in Japan is also seriously concerned. Smart farming based on advances in sensing, robotics, and information and communication technologies is one of the most promising approaches to support efficient, sustainable, and profitable crop production. This paper reviews the need for smart farming and the state-of-the-art in the related research fields such as satellite- and drone-based remote sensing, machine learning, and phenotyping. The caveats, as well as the potential of these technologies, are discussed in detail towards further advances in the technoscience in crop production and their efficient implementation in smart farming.

In this study, we analyzed the relationship between peanuts yields and meteorological conditions using the results of field performance tests and mesh agricultural meteorological data from 1980 to 2020. We found that the pod weight tended to increase when the temperature was high and the sunshine hours were long immediately after flowering. The pod weight of ‘Nakateyutaka’ was more affected by the meteorological condition near the flowering date than that of ‘Chibahandachi’. The pod weight in the standard sowing date (from mid-May to late May) plot was more affected by meteorological conditions than in the late sowing date (in mid-June) plot, and the pod weight in the late sowing date was more affected by climatic conditions at an earlier growing stage than the standard sowing date plot.

We investigated the effects of 4 rates of phosphorus (P) fertilizer application on growth, grain yield and arbuscular mycorrhizal (AM) infection rate of groundnut (Arachis hypogaea L.) grown in acrisols, in which P binds to iron, in Okinawa. Over 4 years, plants were grown with no addition of P (P0), half the standard rate of P (P0.5), the standard rate (P1), and double the rate (P2) using superphosphate. In year 1, there was no significant difference in growth and grain yield between P0 and P2. By the end of year 4, the growth and grain yields of P0 and P0.5 were significantly reduced, but those of P2 were similar to those of P1. We found a significant logarithmic regression relationship between the grain weight and the amount of P uptake of the grain. Grain P uptakes was in a logarithmic regression relationship with several kinds of P contents in soil. The results suggest that groundnut cultivation in acrisols must be 9 g/m² of P fertilization (superphosphate). The AM infection rate of roots did not decrease in P0.5, P1 and P2, even when the P fertility was high, which implied that AM infection strongly affected groundnut growth.

The effects of high temperature on rice taste, grain quality, grain physical properties, apparent amylose content, and gelatinization were evaluated for 3 consecutive years using products grown in plots supplied with hot water as high-temperature stress. In the taste sensory evaluation, the comprehensive evaluation, appearance and taste, and the stickiness decreased while the hardness increased by the high-temperature treatment in all varieties. The surface stickiness decreased by 11%, the surface stickiness to hardness ratio decreased by 10% and the overall stickiness to hardness ratio decreased by 6%, whereas the overall hardness increased by 4%. The apparent amylose content decreased by 10%. The rice flour gelatinization starting temperature increased by 1.2°C, and the peak viscosity and holding viscosity increased by 9% and 5%, respectively. The breakdown increased by 11%. To identify the relevant palatability characteristics, we surveyed the correlation between statistical analysis parameters. The surface stickiness and surface stickiness to hardness ratio showed the highest correlation, suggesting that these parameters are indicators of the lower eating quality of rice during high-temperature treatment. The gelatinization starting temperature also showed a high correlation and was considered a useful indicator of taste degradation due to high-temperature treatment. The surface stickiness to hardness ratio, which showed the highest correlation with hardness in the taste sensory evaluation, was considered an indicator of grain hardness.

We developed a simple method for estimating nutrient index values of ‘Kinumusume’, a recommended rice variety of Okayama Prefecture, from RGB images. ‘Kinumusume’ was grown under different transplanting dates and fertilizer application conditions. Plant length, number of tillers, and SPAD were measured; and, their product value, the nutrient index value, was calculated at around 30, 20, and 10 days before heading. RGB images of the rice plants were taken just before each day’s growth measurement. Sixteen color indices were calculated from the original RGB images (whole area images) and RGB images from which only the rice area was extracted (rice area images), and models were developed to estimate nutrient index values using single regression analysis and 16 machine learning algorithms. Data from different fields were used for model development and accuracy validation. In both single regression and machine learning, rice area images tended to predict nutrient index values with higher accuracy than whole area images. This was considered to be due to the removal of the effect of the image background. The single regression model showed less decrease in accuracy from training data to test data than the machine learning models, which suggested that the single regression model would be superior to machine learning algorithms in that it can capture more general features even from a small set of data.

To stabilize the production of Tamba-Dainagon Adzuki bean, a local specialty crop in Kyoto Prefecture highly valued by consumers, it is necessary to accumulate knowledge on the environmental response of adzuki bean to changes in climate and soil moisture. In this study, we examined the effects of soil drought on the growth and yield of the adzuki bean variety 'Kyoto-Dainagon' at different times from pre-flowering to post-flowering. Soil drought lowered the photosynthetic rate, and decreased the main stem length and node number. Soil drought around the flowering period reduced the number of pods and bean weight, resulting in a decrease in yield. On the other hand, the photosynthetic rate increased after the reduction of soil drought. The reduction of soil drought during the pre- and post-flowering periods was accompanied by an increase in photosynthetic rate, leading to an increase in bean weight. In addition, delayed leaf senescence was observed in adzuki bean that were under a wet condition after the soil had dried out. This is thought to be caused by the excess source of photosynthesis by pod reduction and/or development of root systems under drought due to the recovery of photosynthetic rate under the wet conditions. Thus, it is clear that the yield is reduced and leaf senescence is delayed by soil drought in adzuki bean.

Recently, the yield of soybean in Fukuoka and Saga Prefecture, which are the major soybean producers in northern Kyushu, is decreasing, while rainfall during the growth period is increasing. Soybean yield was significantly and negatively correlated with rainfall during the growth period, which indicates that wet injury negatively affects soybean yield. This study aimed to clarify the effect of a new planting method involving one-step shallow tillage planting using an up-cut rotary in preventing the reduction of soybean yield caused by wet injury in a field in northern Kyushu. Field studies were conducted from 2019 to 2021. Heavy rainfall occurred just after planting in 2019 (mid- to late July), at the early reproductive stage (mid-September) in 2020, and vegetative stage (early to mid-August) in 2021. In 2019, due to defective germination caused by wet injury, the farmer had to re-plant the seedlings after conventional planting (control) , while good seedling establishment was obtained using the new planting method. The yield was 36% lower in the control than when the new planting method was used, due to delay in plant growth caused by later planting in the control. In 2020 and 2021, the duration of over-humidity in soil was approximately twice longer and the yield was 26 and 40%, respectively, lower in the control. On a 3-yr average, the new planting method prevented the reduction of yield caused by wet injury by 34%.

‘Niji-no-kirameki’ is a Japanese cultivar with medium maturity, high-yield, and high-palatability cultivated in the Hokuriku region. The objectives of this study were (1) to clarify the relationship among grain yield, yield components, and palatability, (2) to calculate the shoot nitrogen content at the full heading stage to achieve the optimum spikelet number per square meter (SN) and (3) to determine the proper amount of nitrogen fertilizer as top dressing corresponding to the growth parameters at the panicle neck node differentiation stage in ‘Niji-no-kirameki’. Grain yield (y) had a positive relation with SN and panicle number per square meter (x),represented by a linear regression model. Top dressing of nitrogen fertilizer significantly increased grain yield when compared at the same amount level of nitrogen application. Grain protein content was negatively related with eating quality and grain thickness. Then to achieve a grain yield of at least 700 g m^–2 with SN of 37,500 and grain protein content of less than 6.5% (15% moisture basis), the shoot nitrogen content at the full heading stage was calculated as 13.2 g m^–2, the optimum value of growth index value (plant height cm × number of tillers per square meter × SPAD value) at the panicle neck node differentiation stage was in the range of 18.2-21.9. The amount of nitrogen fertilizer to give this growth index value would be 2-4 g m^–2 of nitrogen fertilizer. The optimum values of growth parameters at key developmental stages for achieving high-yield and high-palatability in ‘Niji-no-kirameki’ in the Hokuriku region were determined.

Although new rice varieties that are resistant to high-temperature damage during rice grain ripening are being cultivated, the time and effort required to create new varieties is a burden. The objective of this study was to establish a method for selecting rice varieties with tolerance to high temperatures at the rice grain ripening stage focusing on the differences in growth responses to high temperature treatments at the seedling stage between varieties bred to be tolerant to high-temperature damage during the rice grain ripening stage and existing varieties. The plants after germination were grown hydroponically in growth chambers for 14 days under daytime and nighttime temperatures of 25°C and 20°C, respectively, in the control and 35°C and 30°C, respectively, in the high-temperature treatment. Compared to the varieties cultivated as resistant varieties, the growth of existing varieties was slow in the control, but this difference was eliminated by the enhanced growth in the high-temperature treatment. The difference in growth between the high-temperature treatment and control was significantly smaller in the varieties cultivated as resistant varieties than in the existing varieties. From these results, it was considered possible to select varieties with resistance to high-temperature damage during the rice grain ripening stage by selecting varieties showing less difference in the growth of the shoot or the root in the high-temperature treatment as compared to the control at the seedling stage.