Study on Soil Desalination Process of Saline-Alkaline Grassland along the Yellow River in Western Inner Mongolia under Subsurface Drainage

: This study aimed to explore the characteristics of water and salt transport in saline-alkali grassland, effectively guide the ecological construction of saline-alkali grassland along the Yellow River in western Inner Mongolia, and provide a theoretical basis and technical support for the ecological restoration of saline-alkali grassland and regional environmentally sustainable development. The desalination process of saline-alkali soil under the condition of subsurface pipe drainage was simulated using an indoor soil tank test. The variation in soil leaching water consumption, the salinity of the leaching ﬁltrate with time, the accumulation of the ﬁltrate, and the desalination rate of the ﬁltrate under the condition of continuous leaching with a 25 mm head were analyzed. At the same time, the variation process of soil conductivity at 0~40 cm (upper layer), 40~80 cm (middle layer), and 80~120 cm (lower layer) was analyzed. Subsurface pipe drainage can reduce the soil salt content, while reducing the control area of the subsurface pipes can accelerate the soil desalination rate, thus improving the quality of saline-alkali soil. In addition, the leaching effect under 30 cm partition spacing was better than that under 35, 40, and 45 cm partition spacing, and the corresponding desalination rate was higher. Under stable continuous leaching with a 25 mm head, the entire leaching process can be divided into three stages: the rapid desalination, moderate desalination, and stable stages. During the desalination process, the upper, middle, and lower layers were desalinated synchronously, and the desalination rate of the upper layer was the highest, followed by the desalination rate of the middle and lower layers.


Introduction
In recent decades, a series of measures have been explored for the management of saline-alkali farmland in China and abroad. With the implementation of environmental protection policies and the improvement in people's awareness, saline-alkali grassland has begun to receive the attention of researchers, and people have begun to control and protect saline-alkali grassland. Soil management measures of saline-alkali grassland are mainly divided into three types: physical measures, chemical measures, and biological measures. Among these, many studies have been conducted on chemical measures and biological measures, which found that the improvement effect of physical measures on saline-alkali grassland is the most direct [1][2][3]. Physical measures can improve the physical and chemical properties of soil, adjust the state of soil water, fertilizer, air, and heat, and create conditions for vegetation restoration. Common physical measures include deep plowing and leveling land. Research on subsurface management is scarce [4][5][6]. Although there are many related studies on field irrigation leaching of soil, there are few related studies using indoor simulated leaching tests, among which the leaching of grassland soil is rare. However, an indoor simulation test is still the main method to accurately analyze the dynamic migration law of water and salt in saline soil [7][8][9].
The soil in different regions is different. The indoor simulation test helps to understand the basic physical and chemical characteristics of the soil and the law of water and salt migration. It provides ideas and references for subsequent field experiments. It also constitutes a system of field experiments and numerical simulation, which can better explain the applicability of the research.
In this manuscript, the soil desalination process of saline-alkali grassland along the Yellow River in western Inner Mongolia under the condition of subsurface pipe drainage was simulated using an indoor soil tank test, and the characteristics of water and salt transport in saline-alkali grassland were explored. This can effectively guide the ecological construction of saline-alkali grassland, and provide a theoretical basis and technical support for ecological restoration of saline-alkali grassland and sustainable development of the regional environment.

Soil and Water for Washing
The soil used in the experiment was taken from saline-alkali grassland of Chengni Village, Shanba Sandaoqiao Town (40 • 49 42.7" N, 106 • 54 28.5" E), Hangjinhou Qi, Inner Mongolia. The initial salt content was 7.38 g/kg, which was consistent with saline soil. The experiment was carried out by leaching and desalting with distilled water. The physical properties and nutrients of the test soil are shown in Tables 1 and 2.

Testing Apparatus
To study the drainage and salt discharge law and the soil desalination effect under different subsurface pipe spacing, an indoor soil bin test was set up. The experimental design was as follows. We prepared a soil tank with a size of 1.5 m × 1.5 m × 1.5 m. A 50 L water tank was set next to it. We used a constant uniform irrigation and maintained a 25 mm head. The device diagram is shown in Figure 1. The buried depth of the buried pipe was 0.8 m, and there were 12 buried pipes. Every three roots formed a dark tube group, and a board with a length of 1.45 m and a width of 1.2 m was set between the dark tube groups as an interval. The spacing between partitions was treated. Four treatments were set up, and the spacing between the left and right partitions was 30, 35, 40, and 45 cm. The treatment numbers were CK, AG1, AG2, and AG3. The pipe material was PVC and the pipe diameter was 25 mm. After the test began, when the water tank was used to supply water to the soil tank, the saline soil (initial moisture content 0.13, salt content 7.855 g/kg) was first loaded into the soil tank. Each 20 cm was vibrated once until the lower part of the buried pipe depth, and then a layer of non-woven fabric was buried in the soil tank. The vibration compaction process was repeated until the soil depth reached 1.2 m. of the buried pipe depth, and then a layer of non-woven fabric was buried in the soil tank. The vibration compaction process was repeated until the soil depth reached 1.2 m.

Test Method and Observation Content
The experiment began at 8:00 on 11 October and ended at 12:00 on 26 October, with a total duration of 365 h. After the filtrate appeared during the test, the filtrate was taken continuously and its conductivity and pH were measured. The sampling interval of the filtrate was initially set to 4 h. During the test, soil samples were taken four times, on 14 October, 18 October, 22 October, and 26 October. When taking soil samples, soil samples were randomly taken around the four buried pipe areas. Soil samples were collected according to 0~40 cm, 40~80 cm, and 80~120 cm layers and put into plastic bags. The soil samples were taken back to the laboratory for moisture content measurement, and then dried and ground, and passed through a 1 mm sieve to remove impurities for the determination of soil pH and total salt content. According to Wang Zunqin's 'Chinese saline soil' classification standard, in semi-humid and semi-arid areas, in terms of salt content, surface soil is classified as follows: less than 0.1% is non-salinized soil, 0.1-0.2% is mildly salinized soil, 0.2-0.4% is moderately salinized soil, 0.4-0.6% is strongly salinized soil, and more than 0.6% or 1.0% is saline soil. To simulate crop growth, the salt content of the upper layer had no salt, the salt content of the middle layer had moderate salt, and the salt content of the lower layer was strong.
Particle composition of soil: a laser particle analyzer was adopted to measure and analyze the soil quality. The soil type in the research area was determined in accordance with the Classification Criteria for Soil Quality in China. Soil pH value: a PHS-3C pH meter was adopted to measure the pH value of the soil filtrate (soil-water ratio was 1:5). Soil salt content: a DDS-11A conductivity meter and DJS-1C conductivity electrode were adopted to measure the soil salt content (soil-water ratio was 1:5). Content of eight major ions: the soil samples and the filtrate were labelled to measure the content of eight major ions. The content of HCO3 − and CO3 2− ions was determined by the dual indicator HCL titration volumetric method; the content of SO4 2− ions was determined by the EDTA volumetric method; the content of CL − ions was determined by the AgNO3 volumetric Mohr method; and the content of Ca 2+ and Mg 2+ ions was determined by the EDTA nitration method.

Test Method and Observation Content
The experiment began at 8:00 on 11 October and ended at 12:00 on 26 October, with a total duration of 365 h. After the filtrate appeared during the test, the filtrate was taken continuously and its conductivity and pH were measured. The sampling interval of the filtrate was initially set to 4 h. During the test, soil samples were taken four times, on 14 October, 18 October, 22 October, and 26 October. When taking soil samples, soil samples were randomly taken around the four buried pipe areas. Soil samples were collected according to 0~40 cm, 40~80 cm, and 80~120 cm layers and put into plastic bags. The soil samples were taken back to the laboratory for moisture content measurement, and then dried and ground, and passed through a 1 mm sieve to remove impurities for the determination of soil pH and total salt content. According to Wang Zunqin's 'Chinese saline soil' classification standard, in semi-humid and semi-arid areas, in terms of salt content, surface soil is classified as follows: less than 0.1% is non-salinized soil, 0.1-0.2% is mildly salinized soil, 0.2-0.4% is moderately salinized soil, 0.4-0.6% is strongly salinized soil, and more than 0.6% or 1.0% is saline soil. To simulate crop growth, the salt content of the upper layer had no salt, the salt content of the middle layer had moderate salt, and the salt content of the lower layer was strong.
Particle composition of soil: a laser particle analyzer was adopted to measure and analyze the soil quality. The soil type in the research area was determined in accordance with the Classification Criteria for Soil Quality in China. Soil pH value: a PHS-3C pH meter was adopted to measure the pH value of the soil filtrate (soil-water ratio was 1:5). Soil salt content: a DDS-11A conductivity meter and DJS-1C conductivity electrode were adopted to measure the soil salt content (soil-water ratio was 1:5). Content of eight major ions: the soil samples and the filtrate were labelled to measure the content of eight major ions. The content of HCO 3 − and CO 3 2− ions was determined by the dual indicator HCL titration volumetric method; the content of SO 4 2− ions was determined by the EDTA volumetric method; the content of CL − ions was determined by the AgNO 3 volumetric Mohr method; and the content of Ca 2+ and Mg 2+ ions was determined by the EDTA nitration method.

Effect of Subsurface Drainage on Soil Moisture Content
The 0-40 cm soil layer was most affected by leaching and evaporation, and the variation range of soil moisture content was the largest. The soil moisture content of the 40-80 cm soil layer was in a steady state, fluctuating between 23.29% and 28.49%. The soil mass water content of the 80-120 cm soil layer was mainly affected by the spacing of the clapboard. The smaller the spacing, the higher the soil mass water content. In this experiment, the soil moisture content was determined by the drying method, and the soil moisture content was measured before irrigation and 3, 7, 11, and 15 days after irrigation.
The profile distribution of soil mass water content in four different treatment areas of CK, AG1, AG2, and AG3 is shown in Figure 2. It can be seen from the figure that the upper and lower soil moisture content in each area changes greatly. This is due to the constant head continuous irrigation and soil water absorption to a certain extent when the pipe began to discharge the filtrate. The soil moisture content in the middle layer was relatively stable and fluctuated less.

Effect of Subsurface Drainage on Soil Moisture Content
The 0-40 cm soil layer was most affected by leaching and evaporation, and the variation range of soil moisture content was the largest. The soil moisture content of the 40-80 cm soil layer was in a steady state, fluctuating between 23.29% and 28.49%. The soil mass water content of the 80-120 cm soil layer was mainly affected by the spacing of the clapboard. The smaller the spacing, the higher the soil mass water content. In this experiment, the soil moisture content was determined by the drying method, and the soil moisture content was measured before irrigation and 3, 7, 11, and 15 days after irrigation.
The profile distribution of soil mass water content in four different treatment areas of CK, AG1, AG2, and AG3 is shown in Figure 2. It can be seen from the figure that the upper and lower soil moisture content in each area changes greatly. This is due to the constant head continuous irrigation and soil water absorption to a certain extent when the pipe began to discharge the filtrate. The soil moisture content in the middle layer was relatively stable and fluctuated less.

Effect of Subsurface Drainage on Soil Total Salt
Studies have shown that subsurface drainage can accelerate soil desalination and reduce the total salt content of each layer of soil. During the test, the total salt content of each layer of soil in the CK area, AG1 area, AG2 area, and AG3 area was monitored. After the start of the experiment, a total of four samples were taken.
It can be seen from Figure 3 that the total salt content of the soil showed a downward trend. Before 14 October, the total salt content showed a significant linear trend and decreased significantly. After this, the total salt content of the upper soil still decreased rapidly, and the rate of decline of the middle soil was significantly slowed. The slowest decline was in the lower soil, and the later stage showed a more stable trend. Finally, on the 26th day, the four areas reached the upper layer without salt, the middle layer was moderate, and the lower layer was severe, at the end of the test.

Effect of Subsurface Drainage on Soil Total Salt
Studies have shown that subsurface drainage can accelerate soil desalination and reduce the total salt content of each layer of soil. During the test, the total salt content of each layer of soil in the CK area, AG1 area, AG2 area, and AG3 area was monitored. After the start of the experiment, a total of four samples were taken.
It can be seen from Figure 3 that the total salt content of the soil showed a downward trend. Before 14 October, the total salt content showed a significant linear trend and decreased significantly. After this, the total salt content of the upper soil still decreased rapidly, and the rate of decline of the middle soil was significantly slowed. The slowest decline was in the lower soil, and the later stage showed a more stable trend. Finally, on the 26th day, the four areas reached the upper layer without salt, the middle layer was moderate, and the lower layer was severe, at the end of the test.  The total salt content of soil in the 0~40 cm layer is closely related to the growth of crops, and reducing the total salt content of the soil plays a decisive role in the growth of crops [10,11]. During the experiment, the changing trend of soil total salt content in the upper layer of the four regions was the same. The early stage showed a rapid desalination process of the upper soil, and the total salt content decreased rapidly before 14 October. After 14 October, the decreasing trend of AG3 salt content slowed. In general, the total salt content of the upper soil decreased gradually after leaching; on 26 October, at the end of the leaching, salinity in four areas in the upper soil decreased, so that the soil became non-saline; this effect was obvious. Compared with the total salt content of the soil before the start of the experiment, the four regions decreased by 94.32%, 94.13%, 92.63%, and 94.66%, respectively. In general, the area with smaller dark tube spacing had a faster leaching speed and lower total salt content. At the end of leaching, it could be seen that the improvement effect of the subsurface pipe on soil salinity in the upper layer of salinealkali soil was significant.
During the monitoring period of the experiment, the changing trend of the total salt content of the middle soil in the four regions of CK, AG1, AG2, and AG3 was quite different from that of the upper soil, and the leaching rate of the upper soil was significantly faster than that of the middle soil. The total salt content decreased rapidly before 14 October (early stage), which was the rapid desalination stage of the middle soil. From 14 October to 22 October, the decreasing trend of total salt in each region slowed. In general, the total salt content of the middle soil decreased gradually after leaching; on 26 October, the total salt content of the upper soil in the four regions decreased, so that the soil became moderately salinized, and the effect was obvious. Compared with the total salt content of the soil before the start of the experiment, the four regions decreased by 75.00%, 76.09%, 70.23%, and 68.52%, respectively. Overall, the smaller the partition spacing area, the faster the leaching, and the less the total salt.
During the experiment, the variation characteristics of the total salt content of the lower soil in the four regions were as follows: the total salt content of the lower soil decreased rapidly before 14 October, which was the rapid desalination stage of the lower soil. After 14 October, the total salt content in each region slowed significantly and steadily. In general, the total salt content of the lower soil decreased gradually after leaching; The total salt content of soil in the 0~40 cm layer is closely related to the growth of crops, and reducing the total salt content of the soil plays a decisive role in the growth of crops [10,11]. During the experiment, the changing trend of soil total salt content in the upper layer of the four regions was the same. The early stage showed a rapid desalination process of the upper soil, and the total salt content decreased rapidly before 14 October. After 14 October, the decreasing trend of AG3 salt content slowed. In general, the total salt content of the upper soil decreased gradually after leaching; on 26 October, at the end of the leaching, salinity in four areas in the upper soil decreased, so that the soil became non-saline; this effect was obvious. Compared with the total salt content of the soil before the start of the experiment, the four regions decreased by 94.32%, 94.13%, 92.63%, and 94.66%, respectively. In general, the area with smaller dark tube spacing had a faster leaching speed and lower total salt content. At the end of leaching, it could be seen that the improvement effect of the subsurface pipe on soil salinity in the upper layer of saline-alkali soil was significant.
During the monitoring period of the experiment, the changing trend of the total salt content of the middle soil in the four regions of CK, AG1, AG2, and AG3 was quite different from that of the upper soil, and the leaching rate of the upper soil was significantly faster than that of the middle soil. The total salt content decreased rapidly before 14 October (early stage), which was the rapid desalination stage of the middle soil. From 14 October to 22 October, the decreasing trend of total salt in each region slowed. In general, the total salt content of the middle soil decreased gradually after leaching; on 26 October, the total salt content of the upper soil in the four regions decreased, so that the soil became moderately salinized, and the effect was obvious. Compared with the total salt content of the soil before the start of the experiment, the four regions decreased by 75.00%, 76.09%, 70.23%, and 68.52%, respectively. Overall, the smaller the partition spacing area, the faster the leaching, and the less the total salt.
During the experiment, the variation characteristics of the total salt content of the lower soil in the four regions were as follows: the total salt content of the lower soil decreased rapidly before 14 October, which was the rapid desalination stage of the lower soil. After 14 October, the total salt content in each region slowed significantly and steadily. In general, the total salt content of the lower soil decreased gradually after leaching; on 26 October, the total salt content of the upper soil in the four regions decreased significantly, but the soil was still severely salinized. The test monitoring results show that the total salt content of the lower soil was greatly affected by the subsurface pipe drainage project. Compared with the total salt content of the soil before the start of the experiment, the four regions were reduced by 63.91%, 54.83%, 58.93%, and 58.39%, respectively. Overall, the area with smaller partition spacing had a faster leaching speed and lower total salt content. At the end of leaching, it could be seen that the improvement effect of the subsurface pipe on soil salinity in saline-alkali soil was significant.

Analysis of Desalination Water Consumption of Subsurface Pipe Drainage
Thirty hours after the start of the leaching desalination test, leachate flowed out from the bottom of CK, AG1, and AG3. After 60 h of the test, filtrate appeared at the bottom of soil column AG2. In the whole experiment, the total amount of leaching water consumed by the water supply capacity bottles above the CK, AG1, AG2, and AG3 soil columns was 300.6, 350.7, 400.8, and 450.9 L, respectively. The cumulative water consumption is shown in Figure 4. on 26 October, the total salt content of the upper soil in the four regions decreased significantly, but the soil was still severely salinized. The test monitoring results show that the total salt content of the lower soil was greatly affected by the subsurface pipe drainage project. Compared with the total salt content of the soil before the start of the experiment, the four regions were reduced by 63.91%, 54.83%, 58.93%, and 58.39%, respectively. Overall, the area with smaller partition spacing had a faster leaching speed and lower total salt content. At the end of leaching, it could be seen that the improvement effect of the subsurface pipe on soil salinity in saline-alkali soil was significant.

Analysis of Desalination Water Consumption of Subsurface Pipe Drainage
Thirty hours after the start of the leaching desalination test, leachate flowed out from the bottom of CK, AG1, and AG3. After 60 h of the test, filtrate appeared at the bottom of soil column AG2. In the whole experiment, the total amount of leaching water consumed by the water supply capacity bottles above the CK, AG1, AG2, and AG3 soil columns was 300.6, 350.7, 400.8, and 450.9 L, respectively. The cumulative water consumption is shown in  It can be seen from Figure 4 that before 50 h, the cumulative water consumption of each treatment was not significantly different, and the overall trend was increasing. The water consumption rate of CK was 0.945 L/h, that of AG1 was 1.103 L/h, that of AG2 was 1.143 L/h, and that of AG3 was 1.500 L/h. This is because AG3 had the largest spacing and the largest control area, which was 1.5 times that of CK, so the water consumption rate was 1.587 times that of CK. The spacing of AG2 was second only to AG3, and the control area was second, at 1.333 times that of CK, so the water consumption rate was 1.209 times that of CK. The spacing of AG1 was larger than that of CK and the control area was larger, at 1.167 times that of CK, and the water consumption rate was 1.167 times that of CK. It can be seen from Figure 4 that before 50 h, the cumulative water consumption of each treatment was not significantly different, and the overall trend was increasing. The water consumption rate of CK was 0.945 L/h, that of AG1 was 1.103 L/h, that of AG2 was 1.143 L/h, and that of AG3 was 1.500 L/h. This is because AG3 had the largest spacing and the largest control area, which was 1.5 times that of CK, so the water consumption rate was 1.587 times that of CK. The spacing of AG2 was second only to AG3, and the control area was second, at 1.333 times that of CK, so the water consumption rate was 1.209 times that of CK. The spacing of AG1 was larger than that of CK and the control area was larger, at 1.167 times that of CK, and the water consumption rate was 1.167 times that of CK.

Analysis of Soil Desalination Rate under Different Treatments
The experiment began at 8:00 on 11 October and ended at 12:00 on 26 October, with a total duration of 365 h. The test ended when the surface salt content reached a low level, and the middle and lower salt content reached a medium level. The desalination rate of different treatments was calculated according to the amount of salt reduction and the time experienced, and the desalination efficiency under the condition of each unit water volume was analyzed by the total water consumption of CK, AG1, AG2, and AG3. Table 3 shows that the desalination rate was faster under the condition of small intervals. In this study, the desalination rate of CK was 1.005 times that of AG1, 1.032 times that of AG2, and 1.597 times that of AG3. Table 3 also shows that the desalination efficiency of CK was 1.173 times that of AG1, 1.376 times that of AG2, and 2.396 times that of AG3. In the process of leaching and desalination of saline soil, when other conditions are the same, it is recommended to use a smaller area for leaching and desalination, because the test results show that the small area treatment can quickly and efficiently leach salt in saline soil.

Analysis of the Change in Filtrate Mineralization
CK, AG1, AG2, and AG3 treatments were applied from the beginning of the test until after 66 h to all outflow filtrate. It can be seen from Figure 5 that from the beginning of the soil column to the end of desalination, the salinity of the filtrate of CK decreased relatively quickly. The salinity of the filtrate of CK at the same time was always smaller than that of the other three treatments. The smaller the spacing between the partitions, the faster the leaching rate. The filtrate salinity of each treatment decreased exponentially with time, and the decreasing rate of filtrate salinity changed from fast to slow and finally stabilized. Because the salt in the soil column moves with the movement of water, the total salt content in the The fitting curve equations of filtrate salinity with time for CK, AG1, AG2, and AG3 are shown in Table 4. In Table 4, t-the time from the start of the leaching test to a certain time, h; y-the salinity of filtrate at a certain time, g/L.
The filtrate salinity of each treatment decreased exponentially with time, and the decreasing rate of filtrate salinity changed from fast to slow and finally stabilized. Because the salt in the soil column moves with the movement of water, the total salt content in the soil column was higher at the beginning of the test, so the filtrate carries more salt. At the beginning of the test, the salinity of the filtrate was higher, and the salinity of the filtrate in the unit time was also reduced. However, with the progress of leaching desalination, the salt content in the soil decreased, the salt carried by the water decreased, the salt gradually deposited to the bottom layer, the leaching efficiency decreased, and the salinity tended to be stable.

Analysis of soil salt ion composition
To study the changes in various ion compositions in the process of leaching and desalination of saline soil, the ion volume fraction in each area of soil column was continuously analyzed, and the changes in ion composition before and after leaching were compared (Table 5). The soil salt content changed significantly before and after soil column leaching. The salt content of soil columns CK, AG1, AG2, and AG3 before leaching was 20.4 times, 19.6 times, 15.1 times, and 21.9 times that after leaching and desalination (taking 0~40 cm soil layer as an example), and the average salt content before leaching was 18.9 times that after soil column desalination. In addition, CO 3 2− and K + were not detected. The content of HCO 3 − increased significantly, and the content of other ion values decreased significantly. Because of the different migration abilities of soil salt ions, the variation in salt ion concentration is different. The changes in soil cations in the four areas treated with different spacing were significantly different. The relative proportions of Na + , Ca 2+ , and Mg 2+ in saline-alkali soil were considered the basic elements of crop growth. The leaching efficiency of different cations with time was: Na + > Mg 2+ > Ca 2+ . The change process of different cations in each soil layer with leaching time was the same, and the content showed a significant downward trend.
The content of Na + in the soil is an important index that directly affects crop growth, soil porosity, and soil salinity [12,13]. Na + had the highest content of cations, and the content of Na + in different soil layers of all treatments decreased gradually with the increase in leaching time. The leaching efficiency of Na + was also the highest in cations. Compared with the Na + content in the soil before leaching, the 0~40 cm soil layer of CK, AG1, AG2, and AG3 decreased by 99.73%, 99.40%, 99%, and 99.54%, respectively; the 40~80 cm soil layer decreased by 79.28%, 80.44%, 74.26%, and 71.10%, respectively; and the 80~120 cm soil layer decreased by 66.98%, 56.66%, 62.21%, and 60.53%, respectively, which indicates that the leaching effect of CK treatment was significantly better than that of AG1, AG2, and AG3. The leaching effect of the small spacing treatment was relatively good. Therefore, the salt leaching effect of the CK treatment was better than that of AG1, AG2, and AG3. The reason for this is that the regional dense subsurface pipe can shorten the salt transport distance; thus, the salt in the soil can more easily discharge with the subsurface pipe. This can speed up the progress of subsurface pipe drainage and salt discharge, and can better solve the problem of severe salinization in saline-alkali land [14][15][16].
The content of Cl − and SO 4 2− in the soil of each spacing treatment decreased after leaching, and the content of HCO 3 − increased. The reasons for this may include two aspects. First, when the content of Cl − and SO 4 2− in the soil is discharged with the subsurface pipe, the carbonate minerals will gradually dissociate HCO 3 − and CO 3 2− to balance the positive and negative charges in the soil. In addition, with the increase in irrigation water, the content of Cl − and SO 4 2− will be discharged in large quantities, and the negative charges in the soil that need to be neutralized will gradually increase, resulting in more and more HCO 3 − and CO 3 2− . The second aspect is that, due to the solubility of CaCO 3 , the decrease in Ca 2+ content will promote CaCO 3 to produce HCO 3 − and CO 3 2− . Analysis of salt ion composition of filtrate To study the changes in various ion compositions in the process of leaching and desalination of saline soil, the volume fraction of ions in the lower filtrate of CK, AG1, AG2, and AG3 was continuously analyzed.
(1) Analysis of filtrate ion change rate The ion content in various treatment filtrates gradually decreased with the leaching process, and the rate of decline was faster. The concentrations of Na + were 10.733, 9.169, 9.698, and 9.673 g/L at 80 h after the start of the experiment, and decreased to 10.733, 9.169, 9.698, and 9.673 g/L at 361 h and 98 h; these values were only 17.09% and 8.37% of those at 29 h. The Na + content at 129 h after the start of the experiment was only 3.21% and 0.53% of that at 29 h. Other ion changes are shown in Table 6. CO 3 2− and K + ions were not detected in the filtrate during the whole experiment. (2) Analysis of the change in ion content in filtrate with leaching time The variation of filtrate ion content with time is shown in Figure 6.

Discussion
According to the change in filtrate ion content and filtrate salinity with time, the w ter and salt dynamics in the test process can be divided into the following stages: (1) Rapid desalination stage. In the early leaching stage, the leaching water dissolv the solid salt content in the upper soil layer, and thus the soil filtrate with high salt a ion content was transported downwards. When filtrate was observed at the bottom of rig (66 h), the entire soil profile was saturated. The changes in ion content in the filtr during the leaching desalination show that under continuous leaching with a 25 mm he the ion content of the filtrate decreased significantly in the first 96 h after the test start As shown in Figure 6, the faster the curve decreases, the faster the ion content in the filtrate decreases. Until the end of the leaching, the ions in the CK filtrate decrease faster than those in the filtrate of other measures. On the whole, the six kinds of ions in each treatment decreased with the same trend and the same curve shape. Taking HCO 3 − as an example, the HCO 3 − content in CK filtrate decreased faster than in the other three treatments. During the 60-120 h period of the experiment, the rate of HCO 3 − content in AG1, AG2, and AG3 began to gradually decrease, and the decreasing rate during 120-240 h was significantly different. The decreasing rate of HCO 3 − content in CK filtrate at 240-360 h was close to that of the other three treatments.

Discussion
According to the change in filtrate ion content and filtrate salinity with time, the water and salt dynamics in the test process can be divided into the following stages: (1) Rapid desalination stage. In the early leaching stage, the leaching water dissolved the solid salt content in the upper soil layer, and thus the soil filtrate with high salt and ion content was transported downwards. When filtrate was observed at the bottom of the rig (66 h), the entire soil profile was saturated. The changes in ion content in the filtrate during the leaching desalination show that under continuous leaching with a 25 mm head, the ion content of the filtrate decreased significantly in the first 96 h after the test started.
(2) Medium-speed desalination stage. With the continuous infiltration of water, the concentration of soil solution in each layer decreased, and the salt solution in each soil layer moved to the next soil layer; at the same time, the layer received the salt solution moving downward from the previous soil layer. At this time, the concentration of soil solution in each soil layer changed, and the salinity and ion content gradually decreased. The decline rate was lower than that in the first stage.
(3) Stable stage. At 250 h after the start of the experiment, the salt in the upper and middle layers gradually deposited and reached a moderate level. As the leaching continued, the total salt content in the upper layer gradually weakened, and soil gradually entered a mild salt or even no salt condition. The middle layer gradually reached a moderate level, and the lower soil was still slowly desalinated.
In the Caofeidian area under the condition of stable flow, Liu Hu divided the leaching stage into the surface soil desalination stage, the salt peak downward stage, and the bottom soil desalination stage in the desalination process of dredger fill [17]. It was considered that one layer of desalination ends and begins the next layer; that is, layered desalination. It was concluded from this study that three-layered soil is desalted synchronously under the condition of steady flow leaching, and the desalination rate is different: upper layer > middle layer > lower layer. When the upper layer becomes salt-free, the lower layer reaches a severe level.
In the rapid desalination stage, the soil salinity and ion concentration in each soil layer decreased rapidly. Because most of the salt contained in the upper layer of the soil column has moved to the next layer, the concentration of the upper soil solution is reduced. The soil layer below the upper layer, in addition to its own salt, also takes the salt from the upper soil layer, so the desalination speed is slower than that of the upper layer. Because most of the salt in the upper layer of the soil column has been moved to the next layer, the concentration of soil solution in the upper layer decreases rapidly. With the entry of salt into the upper soil layer, the desalination rate of the middle and lower soil layers is slowed, so it is slower than that of the upper layer. In the medium-speed desalination stage, because all the salts in the upper and middle soil layers of each treatment moved to the bottom soil layer, the total salt content and ion content in this layer still fluctuated, and due to the accumulation of salt in the upper soil layer, the peak value was higher than that in the upper and middle soil layers. After entering the stable stage, the upper layer entered a mild salt or even no salt condition, the middle layer reached a moderate level, and the desalination rate of the lower soil was slow. At this stage, the salinity and ion content of the filtrate were relatively stable.

Conclusions
1. Subsurface drainage can reduce the degree of soil salinization, and reducing the subsurface control area can accelerate the rate of soil desalination, to achieve the purpose of soil salinization improvement. The leaching effect of 30 cm spacing is better than that of 35, 40 and 45 cm spacing, and the soil desalination rate is faster. Under the condition of small spacing, the desalination rate is faster: the desalination rate of 30 cm spacing is 1.005 times that of 35 cm spacing, 1.032 times that of 40 cm spacing, and 1.597 times that of 45 cm spacing. In addition, the desalination efficiency of 30 cm baffle spacing is 1.173 times that of 35 cm baffle spacing, 1.376 times that of 40 cm baffle spacing, and 2.396 times that of 45 cm baffle spacing.
2. The leaching efficiency of different cations with time is: Na + > Mg 2+ > Ca 2+ ; the decrease in different anions is: Cl − > SO 4 2− > HCO 3 − . The content of ions other HCO 3 − decreased with the increase in leaching time, and the content of HCO3-increased. 3. During the whole leaching process, the content of six kinds of ions in the filtrate of each treatment decreased, showing the same trend, and the shape of the curve of ion content in the filtrate with leaching time was the same. 4. Under the condition of 25 mm head steady flow, the whole leaching process can be divided into three stages: rapid desalination stage, medium-speed desalination stage, and stable stage. In the process of desalination, the upper, middle, and lower layers of soil were desalted synchronously, and the desalination rate was ranked: upper layer > middle layer > lower layer.