Attribution of climate change and human activities to urban water level alterations and factors importance analysis in Central Taihu Basin

Highlights

• LR, RF and SVM were used for regression and SVM performed best.

• The increasing of water level was primarily caused by human activities.

• Human activities contributed the most in the sub-period II (2000–2009).

• Cumulative antecedent precipitation of five days was one of the important climate factors.

• Daily maximum discharge of sluices were the important human activities factors.

Abstract

Attribution of climate change and human activities have been extensively discussed over the past few decades, particularly in urbanization area. However, the relationships among different factors may not be well explained by traditional statistical methods. In this study, we took one of the highly urbanized regions, Central Taihu Basin, as an example. Linear regression (LR), random forest (RF) and support vector machine (SVM) were used for regression and the attributions of climate change and human activities on water level alterations at different scales from 1961 to 2018 were quantified by residual analysis. The regression results indicated that SVM performed best. Water level at each scale showed an increasing trend and human activities were the dominant influence. The altered period was further divided into three sub-periods and human activities contributed the most in the sub-period II (2000–2009). Finally, the importance of thirty-eight factors were quantified by RF based on daily data series from 2008 to 2018. The results showed that cumulative antecedent precipitation (CAP) of five days was one of the important climate factors and daily maximum discharge of sluices were the important human activities factors. The methods and results of this study can help to provide support in flood control.

Introduction

Climate change and human activities are considered to be two major factors behind altering hydrological processes (Dey and Mishra, 2017). Warming temperatures and changes in the amount, intensity, and form of precipitation have a global impact on environment with significant implications on water resources (Arnell, 1999; Kundzewicz et al., 2008; Tramblay et al., 2012). Human activities, such as land use/cover change, hydraulic engineering construction and urbanization, exert multiple pressures on the hydrologic cycle (Shuster et al., 2005). In addition, human activities also interact with climate change (Olson et al., 2008; Berckmans et al., 2019). The superimposed effects of climate change and human activities are complex, especially in urbanization regions. Therefore, quantifying the effects of climate change and human activities on hydrological processes is valuable for strategic water resources management. In this study, climate change refers to the variations of climate factors, such as precipitation, temperature, etc., which can be due to both natural variability and anthropogenic influences. Human activities refer to the human actions that causing regional to local-scale change to the hydrological cycle.

Taihu Basin is one of the typical regions with the fastest urbanization speed and the highest urbanization level in China, which belongs to Jiangsu Province, Zhejiang Province and Shanghai City in administration. The approximately 80% of the total area in the basin is characterized by broad flat geomorphology and lowland plain river networks (Liu et al., 2013). Moreover, due to the abundant marine monsoon precipitation, Taihu Basin is susceptible to threats from flood risk (Zhang et al., 2008). Thus, attribution of climate change and human activities to water level alterations in Taihu plain could provide supportive information for flood operation.

In general, the methods of assessing the impacts of climate change and human activities on hydrological processes mainly include scenarios simulation (Li et al., 2009; Franczyk and Chang, 2009), elasticity analysis (Song et al., 2019) and residual analysis (Wang et al., 2012; Wang, 2014). Scenarios simulation refers to change land use patterns or meteorological data at a time and compare the hydrological model outputs of different scenarios to quantify the influences of climate change and human activities. The advantage of this method is that it is based on stronger physical mechanism and can reflect the inhomogeneity of climate and land use. Nevertheless, it is difficult to apply to larger and complicated basins. Elasticity can be defined as the ratio of proportional change in analysis object like streamflow and water level, and proportional change in relative factors like precipitation, temperature and impervious rate (Ma et al., 2010; Yang and Yang, 2011). This approach can access the impact of representative single factor on analysis object change. However, it is more suitable for estimating the contribution of climate factors, which is limited in the selection of human activities factors. Residual analysis is a common type of quantitative analysis that widely used to quantify the contribution of climate and human activities on the Normalized Difference Vegetation Index (NDVI) (Evans and Geerken, 2004; Zheng et al., 2019) and runoff (Fan et al., 2010). This method is usually based on the assumption that the analysis object is closely correlated with climate factors in one period, i.e., baseline period (Wang et al., 2015). In other words, human activities have no significant impact on the analysis object of analysis in this stage. Analysis object are regressed against climate factors and the resulting model is used to predict the object values for the other stages. Then the residual is the difference between the observed object and predicted object, which indicates that the changes in object are caused by human activities. This approach is simple and can effectively separate the impacts of climate change and human activities.

A large number of hydraulic structures, such as pump, sluices and dikes, have been constructed in Taihu Basin since the 1980s and the total number of hydraulic structures has reached 2468 in the 2000s (Wang et al., 2019a). Due to the complex river network and large amount of hydraulic structures, it is difficult to establish an accurate hydrological model. Consequently, residual analysis was used to quantify the impacts of climate change and human activities on water level in this study. Obviously, the regression model is a key part in residual analysis. However, the previous studies have mostly performed linear model (Li et al., 2017; Liu et al., 2018), which may not explain the relationship between analysis object and multiple factors well. The use of machine learning algorithms can bring significant advantages to both understanding and predicting the climate (Koc and Acar, 2021). Shi et al. (2020) evaluated the contributions of climate factors and human activities to vegetation greenness changes by RF regression with residual analysis. Wang et al. (2019b) distinguished the impacts of climate change and human activities on water level using the multilayer perceptron artificial neural network and residual analysis. In view of the above problems, more nonlinear regression models needed to be applied to obtain more efficient and accurate method.

The main goal of this study is to quantify the impacts of climate change and human activities on water level of Central Taihu Basin. The other objective of this study is to analysis the importance of climate factors (mainly precipitation, temperature and relative humidity) and human activities factors (daily water abstraction and drainage and daily maximum discharge of main sluices), which will be helpful for improving our understanding of important factors and providing reference for dynamic monitoring of flood control.

The water level series of study area from 1961 to 2018 was split into two subseries, i.e., baseline period and altered period, from a break year determined comprehensively by Pettitt test, rank-sum test and Mann-Whitney-Pettitt test. Linear regression and two machine learning models were applied to analyze the relationship between water level and multiple factors under baseline period, and the best one was selected to predict water level during altered period. Then the contributions of two driving factors on water level at monthly, seasonal and annual scale were separated by residual analysis. Finally, the importance of all climate and human activities factors were analyzed by RF based on daily data series from 2008 to 2018.