• The Journal of Engineering Geology
• /
• v.32 no.4
• /
• pp.697-723
• /
• 2022

Data-driven models to predict groundwater levels 30 days in advance were developed for 12 groundwater monitoring stations in the middle-Jeju watershed, Jeju Island. Stacked long short-term memory (stacked-LSTM), a deep learning technique suitable for time series forecasting, was used for model development. Daily time series data from 2001 to 2022 for precipitation, groundwater usage amount, and groundwater level were considered. Various models were proposed that used different combinations of the input data types and varying lengths of previous time series data for each input variable. A general procedure for deep-learning-based model development is suggested based on consideration of the comparative validation results of the tested models. A model using precipitation, groundwater usage amount, and previous groundwater level data as input variables outperformed any model neglecting one or more of these data categories. Using extended sequences of these past data improved the predictions, possibly owing to the long delay time between precipitation and groundwater recharge, which results from the deep groundwater level in Jeju Island. However, limiting the range of considered groundwater usage data that significantly affected the groundwater level fluctuation (rather than using all the groundwater usage data) improved the performance of the predictive model. The developed models can predict the future groundwater level based on the current amount of precipitation and groundwater use. Therefore, the models provide information on the soundness of the aquifer system, which will help to prepare management plans to maintain appropriate groundwater quantities.

• KOREAN JOURNAL OF CROP SCIENCE
• /
• v.27 no.1
• /
• pp.1-10
• /
• 1982

Degree of grain shattering which is of varietal character is an important determinant for the magnitude of field loss of grains during harvest and threshing. Seven Indica \times Japonica progeny varieties and four Japonica varieties were subjected to measurements of tensile strength of grains, degree of grain shattering when panicles were dropped at 1.5m above concrete floor, and moisture content of grains (wet basis) during a period 35 to 63 days after heading. In addition, two varieties were tested for the relation of tensile strength of grains to the magnitude of field loss of grains in actual binder harvest. The 11 varieties differed conspicuously in tensile strength of grains and the degree of grain shattering: the weakest average tensile strength of grains of a variety was about 90g and the strongest about 250g with varying standard deviation of 30 to 60g. Three Indica \times Japonica varieties and one Japonica variety shattered I to 30% of the grains under the falling test. The threshold tensile strength of grains allowing grain shattering was estimated to be 180g on average for a sampling unit of 10 panicles, but only the grains having tensile strength weaker than 98g within the samples shattered. A decrease in average tensile strength by 10g below the threshold value corresponded to an increase of 3 to 5% in grain shattering. Most varieties did not change appreciably the tensile strength of grains and degree of grain shattering with delay in time of harvest and showed a negative correlation between the tensile strength and the moisture content of grains. The average tensile strength of grains was negatively correlated linearly with field loss in binder harvest. The average tensile strength for zero field loss in binder harvest was estimated to be 174g and a decrease in the average tensile strength by 10g corresponded to an increase of 40kg per hectare in field loss of grains. Instead of the average tensile strength of grains, the percentage of grains having tensile strength weaker than 100g is recommended as a criterion for the estimation of field loss of grains during harvesting operations as well as a basis of variety classification for grain shattering, since the standard deviation of tensile strength of grains varies much with variety and time of harvest, and individual grains having tensile strength stronger than 98 did not shatter practically.