SU Chen, WANG Ying, ZHENG Zhao-xian, CHENG Zhong-shuang, CHEN Jiang, HAO Qi-chen

Institute of Hydrogeology and Environmental Geology, Chinese Academy of Geological Sciences, Shijiazhuang 050061, China.

Abstract: The deformation and the drainage of the aquitard is the main concern in the North China Plain (NCP), and the water released from aquitard compaction may be a large portion of the exploited groundwater. The skeletal specific storage of aquitard is the element parameter to the drainage of aquitard, and the undisturbed core samples are the best choice for the measurement of the physical parameters. In this study, the consolidation test was employed to analyze the skeleton specific storage of the clay sample drilled from Hengshui City, and the contribution from the drainage of aquitard to the groundwater exploitation. The results suggest the consolidation test can be utilized to understand the skeletal specific storage of aquitard, which is about 3.92×10-4 m-1 in the Hengshui. The water-saturation content of the aquitard was less than 100%, and the amount of the drainage of the aquitard was about 69%of the volume of land subsidence. The water released from aquitard compaction was about 35% of the groundwater exploitation to the deep aquifers.

Keywords: Hengshui City; North China Plain; Land subsidence; Skeleton specific storage;Aquitard

Introduction

Land subsidence is an environmental geological issue globally. Compression of shallow soil, aquifer deformation and geological settlement are all subsidence types (CHEN C T et al. 2007).The reasons of land subsidence can be divided into exploitation of underground fluid, removal of organic soil, underground mining, soil settlement by soaking, natural deformation of soil, permafrost melting and so on. Excessive exploitation of groundwater, the main cause of regional land subsidence, will result in lower pressure of pore water and higher pressure of clay skeleton (Poland J F, 1984). The original stable structure is broken and partial water is released from the pores in compressed aquitard (Freeze R A and Cherry J A,1970; Devin G et al. 1999). This type of water is a significant part of groundwater exploited in land subsidence areas, and the hydraulic parameter of aquitard closely related to land subsidence has become a research focus (Shave R B, 1998;ZHUANG C et al. 2017a; 2017b; 2017c).

Aquifer specific storage means when water head drops per unit, due to compression of the aquifer’s skeleton and expansion of groundwater,the amount of water released from unit volume of aquifer cylinder (Xue Y Q, 2001). As Terzaghi’s principle of effective stress can apply to both aquitard deformation and aquifer deformation,skeleton specific storage is adopted to show the drainage capacity of aquitard per unit volume in this paper.

Skeleton specific storage can be measured by the ratio of vertical soil deformation volume to total soil thickness, that is, amount of water released from aquitard compaction is equal to the compaction volume of aquitard (Jacob C E, 1940;Terzaghi K, 1943). According to recent domestic researches on undisturbed clay samples, the water-saturation content is less than 100%, which is not exactly the same as ideal state of samples(WEI A H, 2014). Testing some physical and mechanical parameters including compression index and Poisson ratio, or monitoring dynamic changes of groundwater level and aquitard deformation can get hydraulic parameters of aquitard (Jacob C E, 1940; Terzaghi K, 1943;ZHUNG C et al. 2017a; 2017b). But the precondition of all the aforementioned methods is that the pores in aquitard are saturated with water.In fact, compressibility and drainage of soil samples show different trend in various water-saturation content (HU H Y and WANG Z,2007). Numerical simulation of land subsidence,especially regional model, is relatively complicated to get necessary parameters and reliable results (MA Q S and LUO Z J, 2015;ZHOU Y X et al. 2012). The consolidation test is verified that it can be used to study aquitard deformation and its drainage (CAO W B et al.2006). The only difference between pumping test and consolidation test lies in the different way to compress the samples. Despite following Terzaghi’s principle of effective stress, the former one results from drop of water head in aquifer while the latter one is the load from certain device.In consolidation test, relevant parameters of samples can be measured before the test.Deformation and drainage of samples can be recorded at the same time. The consolidation test is a useful way to simulate drainage and deformation of aquitard. However, clay soil samples studied previously are mainly prepared in the laboratory,which could not show its natural characteristic of deformation and drainage, so the analysis on undisturbed soil remains to be done.

Hengshui City, a wide, old and typical land subsidence area in China, is located in the center of“Ji-zao-heng” groundwater depression. Thick aquitard and excessive groundwater exploitation is the cause of the serious land subsidence (XU Y S et al. 2008; WANG M Y, 2016). Main reason is that the discharge of pore water in aquitard increase effective stress of skeleton, resulting in the compression of clay. For the land subsidence in Hengshui City, researches on permeability characteristics of aquitard have been carried out,but few researches focus on its drainage (TIAN L C, 2010; DING G P et al. 2012). The purpose of this paper is to detect the drainage released from aquitard compaction using the consolidation test of undisturbed core samples, analyze the skeleton specific storage of aquitard, and discuss the impacts on the exploited aquifer in Hengshui City.

1 Study area

Hengshui City, located in transition zone between the front part of piedmont alluvialproluvial fan and floodplain in central NCP,overlies a thick Cenozoic sedimentary basin.Quaternary unconsolidated sediment can be divided into 4 aquifer groups. Loose multi-layer aquifer system, confined aquifers as well as normally consolidated or unconsolidated compressible thick aquitard between aquifers are all factors of land subsidence. In addition, overexploitation of groundwater is the main factor for significant decrease of groundwater level, which results in significant land subsidence (HE Q C et al.2006).

This area has a continental semi-arid and semi humid climate. The average annual precipitation is 471 mm, and the average annual evaporation is 1 530 mm in 2001-2005. The Quaternary system in this area is mainly composed of unconsolidated clay and sand with a thickness of 350-500 m.According to the cumulative amount of the land subsidence in the NCP (Fig. 1), the development of land subsidence in Hengshui area is at moderate level, more serious than that in piedmont plain but lighter than that in strand plain. Thus the compaction of aquitard in this area can reflect the average degree of that in NCP.

2 Methods

Samples were taken from borehole by double tube technology. The borehole was drilled to 500 m depth. The lithology and sampling location are shown on Fig. 2. Four samples were used for this experimental study. The samples 1#, 2#, 3#, and 4#were sampled at depth of 20-30, 40-50, 168-187,and 389-396 m, respectively. The samples were wrapped in plastic after taken from borehole, and placed in a sealed plastic box and sent to the laboratory for testing. Consequently, the samples were in an initial condition of physical properties,and the loss of water content was negligible.

Before the experiment, cylinder samples with a length of about 10 cm are cut out, and some physical properties are measured, such as length,water content, water saturation content and so on(Table 1). The experimental equipment is shown in Fig. 3. The equipment is improved on the basis of the WPM3000KN machine. The minimum pressure increment is 1 KN. Besides, two millimeter-scale rulers will be used to measure and check sample deformation. Water released from samples will be measured by measuring glass, and the accuracy is up to 0.1 ml. For shallow samples,the initial pressure is set at 1-5 KN. And the initial pressure is generally more than 10 KN for the deeply-buried samples. Under certain pressure,when there is no significant discharge of water from sample, the consolidation device increases the load at 10-20 KN intervals. When drainage volume is accumulated to a certain amount, the corresponding sample deformation and the consolidation load are recorded.

When the drill core is removed from the underground, the load reduces slightly, so the length of samples will be increased slightly.Deformation and pressure data are revised after experiments. Considering original strata pressure and recovery part after unloading, this study adopts net pressure and net deformation value. Please see specific methods in references (SU C et al. 2014).

3 Discussion

3.1 Water-unsaturated aquitard

When aquitard pores are filled with water, the volume of compressed aquitard is equal to the drainage amount of aquitard. But in this test, the amount of water released from aquitard compaction is greatly lower than the volume of the compressed sample, which indicates that pores in aquitard are not only filled by water. In addition,peculiar smell releasing from samples can further prove that gas and water both exist in undisturbed aquitard.

The water-saturation content of samples mainly ranges from 0.5 to 0.8 (Table 1). The waterunsaturated aquitard is mainly influenced by sedimentary and lithologic characteristics. Gas is the main content of clay soil developed year after year. Due to deposition and crustal movement, clay soil serving as aquitard hinders the direct hydraulic connection between aquifers. It is harder for groundwater to flow in aquitard than sand and gravel layer. The special structure of clay can contain more water, but the gas cannot be discharged absolutely from pores for the weak permeability of clay soil. Even though groundwater leakage happens between aquifers for groundwater exploitation, some holes in aquitard still cannot be totally filled with water.

Fig. 1 The cumulative amount of the land subsidence in the North China Plain to the year of 2003(modified)

Fig. 2 The borehole profile

Table 1 Physical properties of samples in the test

Fig. 3 The diagrams of consolidation equipment.① cylinder barrel; ② plungers; ③ piston; ④ sample chamber; ⑤ ruler; ⑥ base plate; ⑦ ptfe tubing;⑧ measuring glass

3.2 The skeleton specific storage

Load pressure in the consolidation test corresponds to the pressure from the groundwater,so the pressure to samples can be calculated as follows:

So change of water head can be calculated as follows:

Based on calculation method by Riley (1969),skeletal specific storage of water-saturated clay soil can be expressed by:

Formula (3) is the equation for water-saturated clay soil, and in this study the equation for skeletal specific storage of unsaturated clay soil has been revised.

Skis thedrainage of corresponding sample per unit when waterhead decreaseh unit:

The skeleton specific storage of waterunsaturated clay can be calculated as follows:

The results of 4 samples from different depth in the consolidation test are shown in Table 2 and Table 5 below.

Table 2 Data of samples 1# from the testing

Table 3 Data of samples 2# from the testing

Table 4 Data of samples 3# from the testing

Table 5 Data of samples 4# from the testing

The estimated skeleton specific storage of water-unsaturated samples ranges from 1.23×10-4m-1to 7.31×10-4m-1, and the average value is 3.92×10-4m-1(Table 6). In consolidation experiment the load pressure is much greater than its preconsolidation pressure. So the sample deformation in the consolidation test is mainly inelastic compression. Accordingly, inelastic skeleton specific storage of aquitard has been calculated from this study.

The author has participated in the numerical simulation of land subsidence in Tianjin plain area.Through model fitting and correction process, the inelastic skeleton specific storage is 3-6×10-4m-1at Tianjin Plain (LI W Y et al. 2012). The inelastic skeleton specific storage in a site of America is 6-10×10-4m-1, which is more than two orders of magnitude above elastic skeleton specific storage of 10-6m-1(Poland J F, 1984). Hence the data about the inelastic skeletal specific storage of aquitard and the calculation method in this paper are feasible.

3.3 Drainage of aquitard in Hengshui City

Assuming α is the drainage coefficient, namely the ratio of drainage of aquitard with water-unsaturated state to that in saturated state.

According to formulas (3) and (5), we can deduce that:

In this formula, V1is the accumulated amount of drainage of aquitard (L); V2is the accumulated volume of the aquitard compaction (L). In the experiment, as samples compressed, volumes of drainage at different stages are basically lower than corresponding deformation volume, indicating that gas is not expelled completely. Hence α cannot be simply replaced by water saturation content of samples. α (drainage coefficient) of four samples ranges from 0.50 to 0.87 (Table 2-Table 5). The average value of α is 0.69 (Table 6), which means the ratio of drainage volume to the volume of aquitard compaction is 0.69. Supposing that land subsidence is the result of aquitard compaction, we can infer that drainage amount from aquitard compaction accounts for 69% of land subsidence volume.

Table 6 The inelastic skeletal specific storage and drainage coefficient

In this study, the data concerning groundwater exploitation and groundwater over-exploitation in 2001-2004 are collected. In order to analyze groundwater composition and the contribution of drainage resulted from aquitard compaction to deep groundwater, the study discusses groundwater exploitation and land subsidence in 2001-2014.

The groundwater depression was only 2 686 km2in Hengshui City when formed in 1960s.However, the area has reached 6 438 km2in 2000,accounting for 75% area of Hengshui City(ZHANG J X and ZHANG X Z, 2016).Considering that the land subsidence is mainly caused by decline of potentiometric surface of deep aquifers, the distribution of land subsidence is assumed to be approximately equivalent to the area of groundwater depression. Land subsidence developed rapidly in Hengshui City in 2000 with an annual average rate of 75 mm/a. So that the average annual land subsidence volume is estimated to be 4.83×108m3/a in 2001-2004.Aquitard with moderate and high compressibility is thicker and compression of aquitard is the main cause of land subsidence in Hengshui. Hence land subsidence volume is basically equal to volume of aquitard compaction. Furthermore, as the average drainage coefficient is 0.69, drainage amount resulted from aquitard compaction is 3.33×108m3/a.

As annual water supply of deep groundwater in Hengshui City is 9.33×108m3/a in 2001-2004, the drainage amount resulted from aquitard compaction accounts for 35% of deep groundwater withdraw.

Groundwater over-exploitation results in the decline of groundwater level, which breaks the balance between groundwater supply and drainage.Drainage from aquitard can make up such unbalance. From 2001 to 2014, volume of over-exploited groundwater is 3.22×108m3/a,similar to the drainage amount of aquitard of 3.33×108m3/a, so the result is valid in this study.

In central NCP, 1/3 exploited groundwater comes from drainage of aquitard. Such water is crucial to the groundwater resource but the direct result is land subsidence. Therefore, controlling groundwater exploitation of deep aquifers to maintain the potentiometric surface is fundamental to ensure sustainable development of ecological environment.

4 Conclusions

(1) The aquitards in aquifer systems of central NCP is not saturated with water. The volume of drainage resulted from aquitard compaction accounts for 69% volume of land subsidence and deformation. Besides, about 35% of exploited deep groundwater comes from drainage of aquitard.

(2) The consolidation experiment can simulate the change of groundwater level, and can reproduce the process of compression as well as drainage resulted from aquitard compaction. The results of inelastic skeleton specific storage calculated from consolidation experiment shows that inelastic skeletal specific storage of aquitard in central NCP is 1.23×10-4m-1- 7.31×10-4m-1with an average value of 3.92×10-4m-1, providing a reasonable range of original selection in numerical simulation and estimation of subsidence volume.

(3) The experiment adopts single borehole, and the analysis of land subsidence has not considered the stratified contribution of aquitard. Hence there is still much room for improvement in this study.Compaction and drainage of the undisturbed clay soil in different hydrogeological units can be further studied. In-situ monitoring of aquitard deformation at different depth can be adopted to improve the research accuracy and coverage.

Acknowledgements

This study is financially supported by the National Natural Science Foundation of China(NSFC grant No. 41602268 and No. 41702283),and China Geological Survey Project(DD20160238, DD20160311).