Received 6 August 2014; revised 12 September 2014; accepted 24 September 2014

ABSTRACT

Soil salinity at reclaimed tidal land in Korea is highly variable and depending on soil characteris- tics and weather conditions. The objective of this study was to evaluate the temporal changes on soil salinity and the relationships between soil salinity and salinity parameters in reclaimed tidal lands. Soil samples were periodically collected at 0 - 20 cm (surface) and 20 - 40 cm (subsurface) layer at Saemangeum (sandy loam) and Yeongsangang (silty loam) reclaimed tidal land. Electrical conductivity (EC) had a wide range from 0.0 to 34.3 dS∙m⁻¹ during the experimental period. The average soil EC was 6.4 and 3.4 dS∙m⁻¹ at Saemangeum and Yeongsangang reclaimed tidal land, respectively. Soluble sodium concentration had great variations and it was the most important single factor for temporal variations of soil EC regardless of soil textural properties. Calcium con- centration was negatively correlated with soil EC and soluble sodium. Soil exchangeable sodium percentage (ESP) was estimated as a function of soil sodium adsorption ratio (SAR) and a linear regression model (ESP = 6.075 (SAR) + 0.677 for Saemangeum and ESP = 3.925 (SAR) + 0.421 for Yeongsangang reclaimed tidal soil) was suggested for predicting soil ESP from soil SAR. The cha- racteristics of reclaimed tidal lands had different shares of saline and sodic properties during the experimental period. Saemangeum reclaimed tidal soil was highly characterized by sodium, but the effect of sodium on Yeongsangang soil was less pronounced. Our results indicate that persis- tent monitoring and modeling on soil salinity at reclaimed tidal soil is fundamental and the re- sults can provide some useful information for deciding management plans for diverse utilization or to reduce salt damage for stable crop production at reclaimed tidal land.

Keywords:

Reclaimed Tidal Land, Soil Salinity, Temporal Variation, Exchangeable Sodium Percentage, Sodium Adsorption Ratio

1. Introduction

Reclaimed tidal lands that locate mainly in west and south coastal areas of Korean peninsula are approximately 350,000 ha, which occupy about 21% of arable lands for crop production [1] . The importance of reclaimed tidal lands is gradually increasing in Korea because stable food production is fundamental in response to increasing world population and improvement of living standard [2] . However, the crop production at reclaimed tidal lands is limited due to its high soil salinity level [3] .

It is projected that salt damages in reclaimed tidal land is likely to increase with future climate change scenarios such as the rise in sea level and temperature that will subsequently increase evaporation rate and salt accumulation in root zone. It is well reported that salinity level in agricultural soils depends on the variations of many factors such as soil texture, plant growth, quality of irrigation water, hydraulic conductivity, and irrigation system in place [4] - [6] . Therefore, the continuous monitoring on soil salinity can be used to avoid crop yield loss and to manage salinity level in root zone under crop threshold level, which results in diverse utilization of reclaimed tidal land for stable crop production.

The dynamic characteristics of soil salts depend on environmental factors such as nature of soils, the amount and type of soluble salts, and local hydrological conditions [7] [8] . The understanding of predominant salt type is critical because individual ions may affect on soil properties and the possibility that ions may be toxic to crops [9] . Soils with different type and amount of salts show different soil properties, which results in different management practices. The most widely used parameter to evaluate soil salinity is electrical conductivity (EC) that is measuring of salt concentration in soil water [10] . Soil EC is influenced by the type and amount of soluble salts. Soluble salts can easily be leached out of the top surface soil under good drainage conditions. Therefore, management might be based on intercepting water coming into the high EC soils. On the other hand, the presence of excess sodium (Na) can disperse soil particles, which results in poor soil structure with unfavorable physical properties such as reduced aggregate stability, retarded water infiltration, and making plant growth stunted [7] . The content of Na in soils is evaluated either through its exchangeable sodium percentage (ESP) or sodium adsorption ratio (SAR). High Na soil (sodic soil) is required to replace the excessive Na from cation exchange sites with calcium (Ca) supplements. Some studies have been reported that gypsum (CaSO₄・2H₂O) application was effective on salt degradation in reclaimed tidal lands in Korea [11] [12] . The management of soils with high EC and Na is usually more complicated because Ca and magnesium (Mg) salts are preferentially leached away and Na is more slowly leached out and eventually dominates the soil [13] . The management of these soils requires applying Ca supplements followed by desalinization practices of improving the level of water table and soil drainage.

Soil salinity is highly variable by local climate, management practices, and environmental conditions. Reclaimed tidal lands showed the various levels of soil salinity regardless of its establishment time [14] . Therefore, it is necessary to monitor on changes of soil salinity to evaluate and determine the best management practices in reclaimed tidal lands. Although many researchers have assessed the spatial changes of soil salinity at reclaimed tidal lands in Korea [3] [15] , there are limited information available on temporal changes of soil salinity in reclaimed tidal lands to help to decide management practices for reducing crop yield loss. Therefore, the objective of this study was to evaluate the temporal changes of soil salinity and generate soil salinity information to be used a part of any management practices or land use plan of reclaimed tidal land.

2. Materials and Methods

2.1. Study Field Characteristics

Reclaimed tidal lands are salt affected soils that parent material is fluvio-marine deposit soil. Based on soil texture, reclaimed tidal land in Korea can be classified as two categorized soils; 1) sandy loam and 2) silty loam [16] [17] . Therefore, experimental sites were selected at Saemangeum (35˚46'N, 126˚37'E) and Yeongsangang (34˚38'N, 126˚30'E) reclaimed tidal land and their dominant soil series were Munpo (coarse loamy, mixed, non- acid, mesic, TypicFluvaquents) and Poseung (Fine silty, mixed, non-acid, mesic, Typic Endoaquepts), respectively. Both sites have a similar EC values (2.4 and 2.7 dS∙m⁻¹) at soil surface layer on April, 2013. The organic matter (11.6 vs. 2.5 g∙kg⁻¹) and clay content (13% vs. 2%) at Yeongsangang was markedly greater than in Sae- mangeum reclaimed tidal land. The physical and chemical properties of experimental site are shown in Table 1.

The experimental sites have a temperate climate regions with the annual mean temperatures were 12.6˚C and 13.4˚C at Saemangeum and Yeongsangang reclaimed tidal land, respectively. The average long-term annual precipitation was 1250 mm at Saemangeum and 1325 mm at Yeongsangang reclaimed land, on which, about 50% falls between June and August. Under high soil salinity condition bare patches are observed devoid of any vegetation. Some salt tolerant grass species grown in experimental sites were manually removed by hands.

2.2. Sample Collection and Analysis

Three sampling sites were selected at each reclaimed tidal land and three cores of soil samples were taken by a hand spiral auger from surface (0 - 20 cm) and sub-surface (20 - 40 cm) layer weekly from April to November, 2013. The part of collected soil samples were used for mass water content by oven drying at 110˚C and the rest of samples were air-dried, passed through a 2-mm sieve, and then used for the chemical analysis. Soil pH and electrical conductivity (EC) was determined with pH electrode (Beckman PH1560, USA) and conductivity meter (Orion 162A, USA) from 1:5 soil/water suspension. Soil EC of a saturation paste extract was estimated from the EC using 1:5 soil/water suspension and an estimate of soil texture. Organic matter content was measured by Tyurin method (NIAST, 2000). Cation exchange capacity (CEC) was measured by 1N-NH₄OAc at pH 7.0. Particle size distribution was conducted by using ASTM 152H hydrometer after dispersion with 5% sodium hexametaphosphate. Soluble and exchangeable cations were sequentially extracted with distilled water and 1N-NH₄OAc (adjusted pH 7.0) and quantified by inductively coupled plasma-optical emission spectrometer (Varion Vista-MPX, Australia). Sodium adsorption ratio (SAR) and exchangeable sodium percentage (ESP) are given as:

(1)

(2)

Na⁺, Ca²⁺, Mg²⁺ = exchangeable Na⁺, Ca²⁺, and Mg²⁺, respectively, in mmol∙kg⁻¹;

CEC = cation exchangeable capacity, in mmol∙kg⁻¹.

Based on soil EC and ESP values, study soils were classified into four groups: 1) soils with EC ≥ 4 dS∙m⁻¹ and ESP < 15% are designated as saline, 2) soils with EC < 4 dS∙m⁻¹ and ESP ≥ 15% are designated as sodic, 3) soils with EC ≥ 4 dS∙m⁻¹ and ESP ≥ 15% are designated as saline-sodic, and 4) non-salt affected soils [18] . All other soil analysis was conducted based on standard methods by NIAST [19] .

2.3. Statistical Analysis

The data was statistically analyzed using analysis of variance (ANOVA) with the JMP statistical program (SAS institute ver. 5.0, USA). Means between treatments were compared by least significant difference and Tukey- Kramer HSD at 95% level when the F-test was statistically significant.

^†SL, sandy loam; SiL, silty loam.

3. Results and Discussion

During the experimental period, electrical conductivity (EC) had widely ranged from 0.1 to 34.3 dS∙m⁻¹ at Sae- mangeum and from 0.0 to 8.8 dS∙m⁻¹ at Yeongsangang reclaimed tidal land (Table 2). Although EC values at pre-experimental stage showed similar values between surface layers of two locations, the mean EC value of Saemangeum soil was significantly higher (6.4 vs. 3.4 dS∙m⁻¹) than those of Yeongsangang reclaimed tidal soil. The standard deviation of soil EC was significantly greater at Saemangeum, indicating that soil salinity at Sae- mangeum reclaimed tidal land was more fluctuated. This result is supported that soil EC was highly correlated with soluble Na contents in soil and standard deviation of soluble Na at Saemangeum reclaimed tidal soil had considerably higher than those of Yeongsangang reclaimed tidal soil. Although Saemangeum soil showed the less concentration of soluble Ca and Mg compared to Yeongsangang, soil EC value at Saemangeum soil was greater than Yeongsangang soil. Soluble Na concentration of surface layer was substantially greater at Saeman- geum reclaimed tidal land. This result indicates that the temporal variations of soil salinity were much greater at Saemangeum reclaimed tidal soil. The mean sodium adsorption ratio (SAR) and exchangeable sodium percen- tage (ESP) values showed that an identical distribution pattern to EC values at both locations. The mean pH values averaged 7.8 at Saemangeum and 8.1 at Yeongsangang reclaimed tidal land, indicating alkalization level in the study, but the variation in the observed pH was relatively small compared to other salinity parameters.

Soil EC values were significantly influenced by the interaction between sampling site and depth at Saeman- geum (Table 3) and Yeongsangang (Table 4) reclaimed tidal land. Soil EC, soluble Na, SAR and ESP at site 3 of Saemangeum reclaimed tidal land was significantly higher than other two sites. In site 3, soil EC value at surface layer was significantly greater than sub-surface layer, which was attributed to increase in concentration of soluble Na at surface layer. The sub-surface layer showed a similar concentration of soluble Ca and Mg than those of the surface layer, while the concentration of soluble Na at the surface layer were substantially higher at Saemangeum reclaimed tidal soil. These results probably reflect a displacement and accumulation of salts from the lower layer towards soil surface. It is well reported that environmental factors such as soil porosity, quantity of percolation water, nature of the ions present determine the mobility of different ions and their differential spa- tial behavior [8] [20] . The texture of Saemangeum soil was loamy sand and the hydraulic conductivity was much greater compared to Yeongsangang reclaimed tidal soil [21] , which indicate that the upward movement of water and salts was much greater at Saemangeum reclaimed tidal soil. The standard deviations of soluble Na were also much greater, which suggest that soluble Na was the primary ion on the variations of soil EC. Con- trary to Saemangeum soil, the EC value of the sub-surface layer at Yeongsangang reclaimed tidal soil was lower compared to the surface layer (Table 4). The composition of soluble ions had a similar pattern that the concen- tration of soluble Ca and Mg were greater at the surface layer, while the soluble Na concentration was signifi- cantly greater at sub-surface layer than the surface layer in all sites.

Values in parenthesis are values from sub-surface layer (20 - 40 cm). ^†MWC, mass water content; SAR, sodium adsorption ratio; ESP, exchangeable sodium percentage.

^†MWC, mass water content; SAR, sodium adsorption ratio; ESP, exchangeable sodium percentage.

^†MWC, mass water content; SAR, sodium adsorption ratio; ESP, exchangeable sodium percentage.

The mean SAR and ESP values showed an identical distribution pattern to the mean EC values at Yeongsan- gang reclaimed tidal land (Table 4). Many studies have shown that there is a correlation between EC and SAR due to its close correlation between both parameters [8] [22] . The relationship between SAR and ESP of Sae- mangeum and Yeongsangang reclaimed tidal soil is shown in Figure 1. Many researchers have been assessed to predict the relationship between soil ESP and SAR because the determination of soil ESP is necessary to analyze soil cation exchange capacity (CEC), which is time consuming laboratory test [23] [24] . There were linear re- gressions between soil SAR and ESP in ESP = 6.075 (SAR) + 0.677 (R² = 0.947) and ESP = 3.925 (SAR) + 0.421 (R² = 0.895) for Saemangeum and Yeongsangang reclaimed tidal land, respectively. Soil ESP values were much greater compared to soil SAR at both locations. This result is probably due to the characteristics of rec- laimed tidal lands which contain high concentration of soil exchangeable Mg. The previous study suggested that proportionality coefficient from the relationship between ESP and SAR is related to the Gapon equation [25] . The data indicates that the SAR value of Saemangeum soil will result in a 55% more increase in ESP than Yeongsangang soil. These unique relationships between soil ESP and SAR were reasonable since the relation- ship is related to intrinsic soil properties [26] .

Electrical conductivity is influenced by physical and chemical properties of soil including soluble salts, soil water content, bulk density, temperature, and organic matter content [27] . The salt compositions were correlated in an attempt to establish any relationship between soluble and exchangeable cations and their influence on the EC values (Table 5). The results showed that a higher linearity for soluble Na (r = 0.937) and total sum of so- luble cations (r = 0.9391) were obtained from Saemangeum reclaimed tidal land. These results indicate that so- luble cations, especially soluble Na, have the most influencing factor on soil EC. This result was corresponding with the results that Na had the greatest influence on soil EC than other cations [8] . The data showed that K and Mg values were widely dispersed and had a relatively low correlation coefficient value regardless of soluble and exchangeable forms. Even though Ca was not a predominant form of soil salts, there were a negatively correla- tion between exchangeable Ca and other salt compositions. It is reported that Ca ion competes for the exchange sites on soil surface by Na ion so that increasing the Ca concentration reduces the concentration of Na [28] . These results indicate that the supplement of Ca ion may improve soil salinity at reclaimed tidal lands.

The data of this study showed that soil EC and ESP values were highly fluctuated during the experimental pe- riod, which resulted in different shares of saline and sodic properties of experimental soils (Table 6). For exam- ple, the 66.7% and 88.7% of Saemangeum reclaimed tidal soil was salt-affected at surface and sub-surface layer, respectively. Salt-affected Saemangeum soils had characteristics of sodic or saline-sodic soils, indicating that

∑sol. and ∑Exch. denote summation of soluble and exchangeable cations (Ca + K + Mg + Na) concentration, respectively.

^†Normal (EC < 4 dS∙m⁻¹, ESP < 15%), Saline (EC ≥ 4 dS∙m⁻¹, ESP < 15%), Sodic (EC < 4 dS∙m⁻¹, ESP ≥ 15%), Saline-sodic (EC ≥ 4 dS∙m⁻¹, ESP ≥ 15%).

Saemangeum soils were usually associated with high amounts of Na salt. These results were probably due to the characteristics of Saemangeum soil that has high contents of soluble Na, low values of cation exchange capacity and clay content. Contrary to Saemangeum soil, the characteristics of sodic or saline-sodic soils were occasio- nally found in the surface layer at Yeongsangang reclaimed tidal land. The percentage of non-salt affected cha- racteristics at Yeongsangang soil was much greater (63.9% vs. 13.9%) at the surface layer compared to the sub- surface layer. The percentage of normal soils at sub-surface layer was decreased and showed the predominant forms of saline and saline-sodic characteristics at sub-surface layer. The soil with higher levels of Na restricts water-holding capacity, prevent flocculation of soil particles and prevent water penetration [29] . Thus, the less water penetration will become more sodic or saline-sodic properties. The results of these salinity classification showed that Saemangeum and Yeongsangang reclaimed tidal soils had a different characteristic of soil salinity during the growing season.

4. Conclusion

The results of this study showed that there was great correlation between soil EC and soluble sodium content of reclaimed tidal soils regardless of soil texture. Soil ESP can be estimated from soil SAR due to the good linear regressions between soil ESP and SAR at both Saemangeum and Yeongsangang reclaimed tidal soils. Saeman- geum and Yeongsangang soils had different characteristics of saline and/or sodic soils during the growing sea- son. Saemangeum soils were associated with high amounts of sodium, but the effect of sodium on Yeongsan- gang soil was less pronounced. The data of this study indicated that any future management practice or land use plan at Saemangeum and Yeongsangang reclaimed tidal lands should be considered characteristics of soil salin- ity.

Acknowledgements

This study was carried out with the support of “Cooperative Research Program for Agriculture Science & Technology Development (Project No. PJ010258)” Rural Development Administration, Republic of Korea.

References

NOTES

^*Corresponding author.