Determining the change in precipitation is very important in water supply planning, flood control, management of water resources, the sustainability of the ecosystem, and economic development. This study investigated the relationship between three change analysis methods, trend analysis, de-trended fluctuation analysis (DFA), and homogeneity analysis, in precipitation change. These methods were applied to monthly data obtained from Konya Closed Basin in central Turkey. Seasonal data were obtained by collecting monthly precipitation data to form the seasons. As the trend method, Mann–Kendall and Linear Trend methods were used. The linear de-trend method was used as the de-trend method. Standard Normal Homogeneity Tests and Buishand tests were used as homogeneity methods. The models were compared with each other using the critical test method with respect to a 95% of the confidence interval. The comparison, trend, and homogeneity methods gave similar results. According to the homogeneity test, the inhomogeneous stations and the years of change were determined. When trend analysis was applied to these stations, it was determined that the trends are statistically significant. Finally, the question of whether this meaningful trend is due to the deterioration of homogeneity is determined by de-trend analysis. The results indicated that the applied de-trend method showed that statistically significant trends occurred with the deterioration of homogeneity.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Homogeneity, Trend, and De-trend Analysis of Seasonal Precipitation Series of Konya Closed Basin

  • Vahdettin Demir

摘要

Determining the change in precipitation is very important in water supply planning, flood control, management of water resources, the sustainability of the ecosystem, and economic development. This study investigated the relationship between three change analysis methods, trend analysis, de-trended fluctuation analysis (DFA), and homogeneity analysis, in precipitation change. These methods were applied to monthly data obtained from Konya Closed Basin in central Turkey. Seasonal data were obtained by collecting monthly precipitation data to form the seasons. As the trend method, Mann–Kendall and Linear Trend methods were used. The linear de-trend method was used as the de-trend method. Standard Normal Homogeneity Tests and Buishand tests were used as homogeneity methods. The models were compared with each other using the critical test method with respect to a 95% of the confidence interval. The comparison, trend, and homogeneity methods gave similar results. According to the homogeneity test, the inhomogeneous stations and the years of change were determined. When trend analysis was applied to these stations, it was determined that the trends are statistically significant. Finally, the question of whether this meaningful trend is due to the deterioration of homogeneity is determined by de-trend analysis. The results indicated that the applied de-trend method showed that statistically significant trends occurred with the deterioration of homogeneity.