• The Korean Journal of Applied Statistics
• /
• v.34 no.3
• /
• pp.439-447
• /
• 2021

Due to the frequent emergence of global abnormal climates, related studies on meteorological change is being actively proceed. However, the research on trend analysis using weather data accumulated over a long period of time was insufficient. In this study, the trend of temperature time series data accumulated from automated surface observing system (ASOS) for 40 years was analyzed by using a non-parametric analysis method. As a result of the Mann-Kendall test on the annual average temperature and seasonal average temperature time series data in South Korea, it has shown that an upward trend exists. In addition, the result of calculating the Sen's slope, which can determine the degree of tendency before and after the searched change point by applying the Pettitt test, recent data after the fluctuation point confirmed that the tendency of temperature rise was even greater.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.57 no.6
• /
• pp.153-162
• /
• 2015

In this study, we analyzed the trends of water quality along the main stream in Nakdong river basin using the recent data and seasonal Mann-Kendall test. Monthly averaged values of DO, BOD, SS, COD, TN, and TP from 1989 to 2014 for 14 stations (including 2 TMDLs stations) were used in the study. The trend analysis results showed that BOD and TP at most stations has decreasing temporal trend except a few stations while COD and SS showed increasing trend at most stations. Temporal trends in TN at 8 stations were found to be statistically significant and 5 of them showed increasing temporal trend. Temporally averaged BOD, COD, TN and TP were generally increasing as going downstream and the worst water quality were found at Goryeong and Hyunpung station. Overall, water quality of Nakdong river especially in COD, SS, and TN getting worse in time at most stations and as going downstream.

• Journal of the Korean Society of Hazard Mitigation
• /
• v.9 no.6
• /
• pp.17-30
• /
• 2009

In this study, the existence of possible deterministic longterm trend of precipitation amount, monthly maximum precipitation, rain day, the number of rain day greater than 20mm, 30mm, and 80mm was analyzed using the Mann-Kendall rank test and the data from 62 stations between 1905 and 2004 in South Korea. Results indicate that the annual and monthly rainfall amount increases and the number of rain days which have more than 80mm rainfall a day, increases. However the number of rain days decreases. Also, monthly trend analysis of precipitation amount and monthly maximum precipitation increases in Jan., May, Jun., Jul., Aug., and Sep. and they decrease in Mar., Apr., Oct., Nov., and Dec. Monthly trend of the number of rain day greater than 20mm, 30mm, and 80mm increases in Jun., Jul., Aug., and Sep. However results of Mann-Kedall test demonstrated that the ratio of stations, which have meaningful longterm trend in the significance level of 90% and 95%, is very low. It means that the random variability of the analyzed precipitation related data is much greater than their linear increment.

• Journal of Wetlands Research
• /
• v.6 no.4
• /
• pp.35-43
• /
• 2004

The non-parametric Mann-Kendall(MK) statistical test has been widely used to assess the significance of trend in hydrologic time series. The test requires sample data should be serially independent. If sample data is serially correlated, the presence of serial correlation in a time series will affect the test ability for trend analysis. So, we would like to use the modified MK test which uses the effective sample size(ESS) to eliminate the effect of serial correlation in a series. This study investigates the ability of ESS to eliminate the influence of serial correlation of MK test by Monte Carlo simulation and by real series. As the results, MK test shows the increase of trend rate as the serial correlation is increased but the modified MK test shows ESS can eliminate the serial correlation for trend analysis. Therefore we confirmed the modified MK test is a very useful tool for trend analysis.

• Journal of Korea Water Resources Association
• /
• v.46 no.3
• /
• pp.219-227
• /
• 2013

Time series data including change, jump, trend and periodicity generally have nonstationarity. Especially, various methods have been proposed to identify the trend about hydrological time series data. However, among various methods, evaluation about capability of each trend test has not been done a lot. Even for the same data, each method may show the different result. In this study, the simulation was performed for identification about the changes in trend analysis according to the statistical characteristics and the capability in the trend analysis. For this purpose, power test for the trend analysis is conducted using Men-Kendall test, Hotelling-Pabst test, t test and Sen test according to the slope, sample size, standard deviation and significance level. As a result, t test has higher statistical power than the others, while Mann-Kendall, Hotelling-Pabst, and Sen tests were similar results.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2016.05a
• /
• pp.232-232
• /
• 2016

주어진 시계열 자료의 경향성을 분석하고 판별하는 것은 수문 자료의 분석에서 가장 우선적으로 수행되어야 할 절차이며 경향성의 유무에 따라 자료를 분석하는 방법이 달라지게 되므로 매우 중요한 부분이다. 일반적으로 국내에서 주로 사용되는 수문 시계열 자료의 경향성 분석 방법으로는 비매개변수적인 방법인 Mann-Kendall test, Spearman's rho test, Hotelling Pabst test, Sentest 등이 있으며 그 중에서도 국내외 수문 자료의 경향성 분석에는 비교적 높은 기각력을 보이는 Mann-Kendall test가 주된 방법으로 활용되어 오고 있다. Mann-Kendall test는 통계적 유의성을 바탕으로 한 경향성 판별 방법으로 시계열 자료 내에 존재하는 경향성의 형태를 분석하여 경향성 유무를 판별하는 것에는 한계가 있다. 경험적 모드분해법을 활용한 경향성 분석 방법은 체거름 과정을 통하여 주어진 시계열 자료를 내재모드함수로 분해한 후, 추출된 모든 요소를 제거하고 남은 잔여값의 형태를 이용하여 경향성 유무를 판별하는 방법으로 자료에 내재된 경향성의 형태를 확인할 수 있는 장점을 가지고 있다. 본 연구에서는 이러한 경험적 모드분해법을 이용한 경향성 분석 방법을 소개하고, 모의를 통한 시계열 자료를 이용하여 경향성 분석에 적용한 후 기존에 사용되어온 Mann-Kendall test와의 비교를 통해 적용성을 평가하였다.

• Journal of Environmental Impact Assessment
• /
• v.4 no.2
• /
• pp.93-103
• /
• 1995

This study was carried out to suggest the best statistical test to analyze the trend in monthly water quality data. Traditional parametric tests such as t-test and regression analysis are based on the assumption that the underlying population has a normal distribution and regression analysis additionally assumes that residual errors are independent. Analyzing 9-years monthly COD data collected at Paldang in Han River, the underlying population was found to be neither normal nor independent. Therefore parametric tests are invalid for trend detection. Four Kinds of nonparametric statistical tests, such as Run Test, Daniel test, Mann-Kendall test, and Time Series Residual Analysis were applied to analyze the trend in the COD data, Daniel test and Mann-Kendall test indicated upward trend in COD data. The best nonparametric test was suggested to be Daniel test, which is simple in computation and easy to understand the intuitive meaning.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2017.05a
• /
• pp.193-193
• /
• 2017

Trend analysis can enhance our knowledge of the dominant processes in the area and contribute to the analysis of future climate projections. The results of previous studies in South Korea showed that southeast regions of Korea had the highest value of evapotranspiration. Thereby, it is of interest to determine the trend analysis in hydrological variables in this area. In this study, the recent 35 year trends of precipitation, reference evapotranspiration, and aridity index in monthly and annual time scale will be analyzed over three stations (Pohang, Daegu, and Pusan) of southeast Korea. After removing the significant Lag-1 serial correlation effect by pre-whitening, non-parametric statistical Mann-Kendall test was used to detect the trends. Also, the slope of trend of the Mann-Kendall test was determined by using Theil-Sen's estimator. The results of the trend analysis of reference evapotranspiration on the annual scale showed the increasing trend for the three mentioned stations, with significant increasing trend for Pusan station. The results obtained from this research can guide development if water management practices and cropping systems in the area that rely on this weather stations. The approaches use and the models fitted in this study can serve as a demonstration of how a time series trend can be analyzed.

• Journal of Korea Water Resources Association
• /
• v.47 no.7
• /
• pp.643-656
• /
• 2014

The homogeneity analysis of temporal (monthly, seasonal and annual) climate aridity index trend was accomplished for 43 climate measurement stations in South Korea. Furthermore, 43 stations were grouped into 9 different regions and the temporal and regional homogeneity of climate aridity index trends in each region and entire 9 regions were analyzed. For analysis, monthly, seasonal and annual climate aridity indexes of 43 study stations were estimated using precipitation and potential evapotranspiration calculated from FAO Penman-Monteith equation. The Mann-Kendall statistical test for significant trend was accomplished using the estimated climate aridity indexes and the results of trend test (Z scores) were used to analyze the temporal and regional homogeneity of climate aridity index trends. The study results showed the temporal and regional homogeneity of climate aridity index trends for individual and entire 9 regions. However, the homogeneity and the extent of aridity index trend showed different patterns temporally and regionally.

• Journal of Soil and Groundwater Environment
• /
• v.14 no.5
• /
• pp.41-50
• /
• 2009

Water levels in groundwater monitoring wells of Jeju Island were analyzed using parametric and non-parametric trend analyses. Number of used monitoring wells in the analysis are 94 among totally 106 monitoring wells and the monitoring period is greater than single year, from 2001 to 2009. For the trend analysis, both parametric (linear regression) and nonparametric (Mann-Kendall trend test and Sen's trend test) methods were adopted. Results of the linear regression analysis on daily basis indicated that about 58.5% of the monitoring wells showed a decreasing trend, and analysis using monthly median indicated that about 79.8% showed a decreasing trend. The Mann-Kendall trend test and Sen's trend test with monthly median values in confidence levels of 95% and 99% showed the same analysis results. In confidence level of 95%, 32% were decreased, 3% were increased and the remains showed no trend. However, in confidence level of 99%, 16% were decreased, 2% were increased and the remains showed no trend. The largest decline rates of water levels were detected mainly at the coast of the northwestern and southwestern parts, which is expected to closely related to the increased pumping in the urban area and tourist resort.