Constructing Italian ryegrass yield prediction model based on climatic data by locations in South Korea

Abstract

A yield prediction model for Italian ryegrass (IRG) was constructed based on climatic data by locations in South Korea using a general linear model. The sample size of the final dataset was 312 during 25 years. The forage crop and climatic data were collected from the reports of two national research projects on forage crops and Korean meteorological administration, respectively. Five optimal climatic variables were selected through the stepwise multiple regression analysis with dry matter yield (DMY) as the response variable. Subsequently, three climatic variables were selected after considering the interpretability of the five variables. The three selected climatic variables were spring accumulated temperature, mean temperature in January and spring rainfall days. Then, the yield prediction model was constructed based on these three climatic variables using general linear model with the cultivated locations as dummy variables. The model constructed in this research could explain 73.6% of variation in DMY of IRG. The goodness‐of‐fit of the model was tested through residual diagnostics and 10‐fold cross‐validation. For climatic variables, the high partial eta squared value of spring accumulated temperature and spring rainfall may reflect the growth characteristics that spring is the main growing period for IRG and IRG has strong waterlogging tolerance and weak drought tolerance. The results may also support the possibility to sow IRG in the subsequent spring if autumnal seeding was missed in South Korea.

Yield modeling for prediction of regional whole‐crop barley productivity

“…According to the year and cultivated location in the data, raw meteorological data such as daily mean, maximum, and minimum temperatures, daily precipitation and daily sunshine duration were collected from the database of the Korea Meteorological Administration. Afterward, 12 climatic variables were generated as shown in Table referring to the previous research (Kim et al, ; Peng et al, , Peng, Kim, Kim et al, ) and subsequently combined into the whole‐crop barley production data. The final data set with a sample size of 290 containing dry matter yield (DMY) values, eight cultivated locations (Figure ), and the 12 climatic variables was applied in the following analyses after eliminating 12 data points with missing values and 14 outliers.…”

“…Analyzing the effects of climatic factors on forage crop production has provided aid for determination of agricultural policies, cultivation practices and decision‐making (Martre, Yin, & Ewert, ). The regional yield prediction modeling based on climatic information on forage crops is considered as requisite to be performed responding to the requirement of the local producers to get to know the crop yield in advance and arrange their production and trade schedules (Basso, Cammarano, & Carfagna, ; Peng, Kim, Kim et al, ).…”

“…Several methods, such as remote sensing, dynamic growth model and empirical crop yield prediction modeling, had been developed to predict the crop yields (Basso, ; Cao, White, & Wang, ). As mentioned by Peng, Kim, Kim et al (), not like food and cash crops, the cultivation of forage crops in South Korea is geographically scattered and the area of cultivation site is not large; meanwhile, the data recording status of cultivation information of forage crops was not sufficient. These situations lead to the difficulties to apply remote sensing or dynamic growth models in South Korea to aid local producers to predict the yield of forage crops.…”

“…Kim, Jeon, Sung, and Kim () examined the path relationship between whole ‐ crop barley yield and climate factors such as temperature, precipitation and sunshine duration using a Bayesian approach of structural equation model and found that the yield of whole ‐ crop barley was affected by spring temperature directly and affected by spring rainfall indirectly. Furthermore, Peng et al, (); Peng, Kim, Kim et al, (); Peng, Kim, Jo et al, () reported several studies using general linear model to construct the yield predictive models of Italian ryegrass and forage rye. However, few studies were reported on crop ‐ weather models for whole‐crop barley in South Korea.…”

Yield Prediction Modeling for Sorghum–Sudangrass Hybrid Based on Climatic, Soil, and Cultivar Data in the Republic of Korea

“…Similar studies were also carried out in economic plants, such as Chinese cabbage [21] and apple [22], in this region. For forage crops cultivated in this region, yield modeling studies based on climatic data were carried out on whole crop rye [23], Italian ryegrass [24], and whole crop maize [25]. However, no research expounding climatic, soil, and cultivar data-based yield modeling of forage crops cultivated in the southern area of the Korean Peninsula was reported.…”