Ⅰ. INTRODUCTION
According to climate change reports, the global mean temperature has increased by approximately 0.75°C over the past 100 years, and is projected to increase by 2.8–4.8°C by the end of the 21st century (Jung et al., 2019). Such temperature increases and the rise in abnormal climate events are expected to negatively affect domestic cool-season grass productivity (Shin, 2025). Representative species mainly used for pasture establishment in Korea include orchardgrass and tall fescue, and these species are widely utilized in domestic forage production based on their excellent cold tolerance, regrowth capacity, and high yield (Hwang et al., 2016;Ji et al., 2020). Recently, the National Institute of Animal Science conducted surveys on optimal cultivation areas and changes in yield for forage crops, along with Climate Impact Vulnerability Assessments, to ensure a stable forage supply for livestock farms. Specifically, a basic Climate Impact Vulnerability Assessment was conducted for both summer and winter forage crops in the study titled “Survey on the Productivity and Impact/Vulnerability Assessment of Grasses and Forage Crops under Climate Change (Phase 1)” (Jung et al., 2024). However, studies and reports on cool-season grasses remain relatively scarce compared with major staple crops such as rice, and productivity-related literature and report data also show large variability depending on region and cultivation timing (Kim et al., 2023;Shin et al., 2023;Shin, 2025). In addition, there is a lack of sufficient validation regarding whether the climate exposure indicators utilized in previous studies are applicable to actual field conditions. Therefore, in this study, we re-examined the relationship between climate factors and productivity by incorporating field experimental results conducted over two years in Jinju and Jangheung into the cool-season grass productivity data and literature review data obtained from the Phase 1 study. Through correlation analyses, we selected climate exposure indicators highly related to cool-season grass productivity, and validated these analysis results by comparing them with the actual yield from the field experiments. Furthermore, based on these findings, we aimed to explore the feasibility of supplementing and refining the previously developed climate exposure indicators.
Ⅱ. MATERIALS AND METHODS
1. Field Experiments on Cool-Season Grasses
This experiment was conducted over two years, during the 2017-2019 growing seasons, at the affiliated animal farm of Gyeongsang National University in Jinju and an experimental field in Jangheung, Jeollanam-do. The field experiments in these two regions were conducted to provide comparative empirical data to validate the correlation analysis results derived from the nationwide dataset. The experimental plot area was 6 m2 (2 × 3 m). The forage species used were tall fescue cv. Greenmaster and orchardgrass cv. Onnuri 2, which were drill-seeded at a seeding rate of 30 kg/ha each. Sowing was performed on September 29, 2017, and October 4, 2018, in the Jinju experimental field, and on September 28, 2017, and October 18, 2018, in the Jangheung experimental field. The fertilizer application rate was 140-120-120 kg/ha of N-P-K. For nitrogen, 30% was applied as a basal dressing and 70% as a top dressing after overwintering. Phosphorus and potassium were applied at 50% as a basal dressing at sowing and 50% as a top dressing after overwintering, respectively. No additional irrigation was provided during the experimental period, and harvesting and growth surveys were conducted on May 25, 2018, and May 28, 2019, in Jinju, and on May 29, 2018, and May 30, 2019, in Jangheung. The fresh yield was measured in each experimental plot. Subsequently, the samples were dried at 65°C for over 72 hours to determine the dry matter percentage. The dry matter yield was calculated by multiplying the fresh yield by the dry matter percentage.
2. Collection of Cool-Season Grass Productivity and Climate Data
Nationwide cool-season grass productivity data from 1993 to 2021 were provided by the Forage Production System Division of the National Institute of Animal Science, and data from 2021 to 2024 were collected through a literature review. A correlation analysis was conducted between climate factors and productivity of cool-season grasses using nationwide productivity data from 1993 to 2024 collected through literature search and the National Institute of Animal Science (n=542), along with productivity data from field experiments conducted in Jinju and Jangheung over two years (n=15; total n=557). Nationwide climate data were obtained through the Automated Synoptic Observing System (ASOS) of the Korea Meteorological Administration. If a weather station was not located in the corresponding region, data from the nearest station within a 20 km radius were used. The drought score represents the total count of 10 consecutive days with no rainfall.
3. Statistical Analysis
Considering the productivity data based on the field experiments and published literature as individual samples, their relationships with climate factors were analyzed using Pearson's product-moment correlation coefficient. A one-way analysis of variance (ANOVA) was conducted to test the significance of the data, and differences between treatment means were subjected to post-hoc analysis using Duncan's multiple range test at a 5% significance level (p<0.05). All statistical analyses were performed using IBM SPSS Statistics (IBM SPSS Statistics for Windows, Version 25.0, Armonk, NY, USA).
Ⅲ. RESULTS AND DISCUSSION
Different letters within the same column indicate significant differences (p<0.05). The significance test was conducted separately for each cool-season grass species (tall fescue and orchardgrass).
1. Comparison of Correlation Analysis and Field Experimental Results of Forage Crops
In this study, to supplement and refine the previously developed Climate Impact Vulnerability Assessment indicators, we examined whether the results of the newly conducted correlation analysis between nationwide cool-season grass productivity and climate factors (Table 1) are applicable to actual cultivation environments. To this end, data from cool-season grass field experiments conducted over two years in Jinju and Jangheung (Table 2) were cross-compared with the correlation analysis results. In the second year, orchardgrass in Jangheung showed significantly higher plant height, fresh yield, and dry matter yield compared to that in Jinju, indicating growth differences between the regions. Regarding the climate conditions in the second year, Jangheung recorded higher mean temperatures (Table 3) and mean minimum temperatures (Table 5) from November to January, as well as higher GDD in November and December (Table 6) compared to Jinju, which subsequently affected the orchardgrass yield. In a comparison by year for each region, the mean temperatures (Table 3), mean minimum temperatures (Table 5), and mean maximum temperatures (Table 4) for November, December, and January were higher in the second year than in the first year for both regions. These temperature conditions likely contributed to the increased fresh yield of tall fescue observed in the second year. The correlation analysis results between cool-season grass productivity and climate factors also showed that the mean temperature and mean minimum temperature from November to March had a significant positive correlation with productivity, and the mean maximum temperature from December to March also showed a significant positive correlation with productivity. Furthermore, GDD in November and from January to March exhibited a significant positive correlation with productivity. This is consistent with the analysis by Kim et al. (2023) that mean temperature and mean minimum temperature are of great importance in determining orchardgrass productivity, and is in line with the report by Isaacson et al. (2024) that temperature increases significantly increase the orchardgrass dry matter yield at the first harvest.
According to the field experimental results, the plant height and fresh yield of tall fescue and orchardgrass cultivated in both Jinju and Jangheung in the second year significantly increased compared to the first year (Table 2). These results are attributed to the relatively higher cumulative precipitation from September to December, as well as in February and May, in the second year compared to the first year in both regions, and this difference in precipitation conditions is considered the cause that induced the increases in plant height and fresh yield of the two species (Table 7). This is in the same vein as the findings of Kim et al. (2023) that precipitation is important for productivity. However, while the dry matter yield of tall fescue in Jinju and Jangheung, and orchardgrass in Jangheung significantly increased compared to the previous year, the dry matter yield of orchardgrass in Jinju showed no significant difference between the years (Table 2). This difference in orchardgrass growth is considered to stem from differences in precipitation amount and the number of rainy days in April, immediately prior to the harvest period (Table 7; Table 8). In April of the second year in Jangheung, the precipitation was 78.5 mm with 11 rainy days, distributed evenly throughout the month without concentration. Conversely, in the second year in Jinju, 114.6 mm of precipitation was concentrated over 7 days; consequently, it is believed that orchardgrass, which has low waterlogging tolerance, suffered from waterlogging damage, leading to a decrease in dry matter yield (Table 7; Table 8). On the other hand, for tall fescue, the dry matter yield significantly increased even under the concentrated precipitation conditions in the second year in Jinju, which is considered to be due to the high waterlogging tolerance of tall fescue (Table 2) (Mui et al., 2021). In addition, the correlation analysis results between cool-season grass productivity and climate factors in this study revealed that the number of rainy days in July had a significant negative correlation with productivity (Table 1). This is considered a result of the summer monsoon hindering the productivity of cool-season grasses (Min et al., 2025). Although there is a difference in the specific timing, these findings are similar to the study by Shin et al. (2023) conducted in Phase 1, which reported a negative correlation with the number of rainy days in August, indicating that frequent rainfall during the summer negatively affects forage crop productivity.
2. Derivation of Climate Exposure Indicators
By cross-comparing the correlation analysis results between nationwide cool-season grass productivity and climate factors with the field experimental results, we selected the following climate exposure indicators (Table 10). Each indicator was selected from climate factors that showed significant p-values and had a correlation coefficient of 0.35 or higher, or -0.35 or lower (Gignac and Szodorai, 2016;Ayed et al., 2021). As a result, the mean temperatures for January, February, March, November, and December were selected. The mean maximum temperatures for January, February, March, and December were selected. The mean minimum temperatures for January, February, March, November, and December were selected. GDD for January, February, March, and November were selected. The number of rainy days in July was selected.
Ⅳ. CONCLUSION
In this study, to examine the field applicability of previously developed Climate Impact Vulnerability Assessment indicators, we compared and analyzed the correlation analysis results between nationwide cool-season grass productivity data and climate data alongside the field experimental results of cool-season grasses. The analysis revealed that cool-season grass productivity exhibited a positive correlation with winter and spring temperatures as well as GDD. Furthermore, it showed a negative correlation with the number of rainy days in July. These results suggest that climate exposure indicators can be utilized to explain the productivity changes and climate responses of cool-season grasses. However, since correlation analysis has limitations in merely presenting associations between variables, it is necessary to continuously refine the explanatory and predictive power of the indicators through further data accumulation and analysis in the future.
