Short Term Effects of Revegetation on Labile Carbon and Available Nutrients of Sodic Soils in Northeast China

In response to land degradation and the decline of farmers’ income, some low quality croplands were converted to forage or grassland in Northeast China. However, it is unclear how such land use conversions influence soil nutrients. The primary objective of this study was to investigate the influences of short term conversion of cropland to alfalfa forage, monoculture Leymus chinensis grassland, monoculture Leymus chinensis grassland for hay, and successional regrowth grassland on the labile carbon and available nutrients of saline sodic soils in northeastern China. Soil labile oxidizable carbon and three soil available nutrients (available nitrogen, available phosphorus, and available potassium) were determined at the 0–50 cm depth in the five land uses. Results showed that the treatments of alfalfa forage, monoculture grassland, monoculture grassland for hay, and successional regrowth grassland increased the soil labile oxidizable carbon contents (by 32%, 28%, 15%, and 32%, respectively) and decreased the available nitrogen contents (by 15%, 19%, 34%, and 27%, respectively) in the 0–50 cm depth compared with cropland, while the differences in the contents of available phosphorus and available potassium were less pronounced. No significant differences in stratification ratios of soil labile carbon and available nutrients, the geometric means of soil labile carbon and available nutrients, and the sum scores of soil labile carbon and available nutrients were observed among the five land use treatments except the stratification ratio of 0–10/20–30 cm for available phosphorus and the values of the sum scores of soil labile carbon and available nutrients in the 0–10 cm depth. These findings suggest that short term conversions of cropland to revegetation have limited influences on the soil labile carbon and available nutrients of sodic soils in northeastern China.


Introduction
The structure, diversity, and production capacity of terrestrial ecosystems are strongly linked to the availability of soil nutrients, such as nitrogen, phosphorus, potassium, and soil organic carbon [1,2]. However, the soil labile carbon and soil nutrients' availability in terrestrial ecosystems are usually influenced by various direct and indirect soil disturbances [3,4]. Land use conversions are major drivers of changes in soil labile carbon and soil nutrient availability, resulting in the degradation of soil ecosystem services (nutrient cycle, water conservation, pollution purification, etc.) and global Land 2020, 9, 10 3 of 14 after conversion from cropland to alfalfa forage, monoculture Leymus chinensis grassland, monoculture Leymus chinensis grassland for hay, and successional regrowth grassland and examine whether short term revegetation could improve the LOC and soil available nutrients in the regions where the soils were not suitable for planting crops in northeastern China.

Study Area
The research was conducted in the Songnen plain located at the Grassland Farming and Ecological Research Station (123 • 31 E, 44 • 33 N) ( Figure 1). The terrain surrounding the study area is relatively flat, and the altitude is approximately 145 m above sea level. The study area has a temperate, semiarid continental climate. The average annual temperature is 5.9 • C, and the mean annual precipitation is 427 mm . The soil is classified as Solonetz in the World Reference Base for Soil Resources with a soil texture of 22% sand, 33% silt, and 45% clay [25]. The main vegetation consists of perennial herbs such as Leymus chinensis and Puccinellia tenuiflora. Besides, some therophytes such as Chloris virgata and Suaeda heteroptera grow in the areas with higher soil pH and poor soil quality [26]. research was to investigate the changes in the contents of labile oxidizable carbon (LOC), AN, AP, and AK after conversion from cropland to alfalfa forage, monoculture Leymus chinensis grassland, monoculture Leymus chinensis grassland for hay, and successional regrowth grassland and examine whether short term revegetation could improve the LOC and soil available nutrients in the regions where the soils were not suitable for planting crops in northeastern China.

Study Area
The research was conducted in the Songnen plain located at the Grassland Farming and Ecological Research Station (123°31′ E, 44°33′ N) (Figure 1). The terrain surrounding the study area is relatively flat, and the altitude is approximately 145 m above sea level. The study area has a temperate, semiarid continental climate. The average annual temperature is 5.9 °C, and the mean annual precipitation is 427 mm . The soil is classified as Solonetz in the World Reference Base for Soil Resources with a soil texture of 22% sand, 33% silt, and 45% clay [25]. The main vegetation consists of perennial herbs such as Leymus chinensis and Puccinellia tenuiflora. Besides, some therophytes such as Chloris virgata and Suaeda heteroptera grow in the areas with higher soil pH and poor soil quality [26].

Experimental Design
This experiment was organized as a completed block design with five land use treatments. In early May 2011, four adjacent blocks (each 60 × 50 m, 2 m buffer between the blocks) in the study area based on similar land use history were identified. Before this experiment (2004)(2005)(2006)(2007)(2008)(2009)(2010), farmers grew rain fed maize (Zea mays L.) and sunflower (Helianthus annuus) in these blocks, following the traditional planting practices in Northeast China, which consists of plowing the soil down to 20 cm depth and applying 50-96 kg N ha −1 , 20-45 kg P ha −1 , and 15-45 kg K ha −1 fertilizers into the soils. The soil properties in these four blocks were homogeneous due to the continuous plowing. The five land use treatments consisted of corn cropland (corn, used as an indication of how the revegetation influences the soils in this study), alfalfa perennial forage land (alfalfa), monoculture grassland of Leymus chinensis (MLG), monoculture grassland of Leymus chinensis for hay (Mowing) once a year (MLG + M), and successional regrowth grassland (SRG) (Figure 2). Leymus chinensis is the native vegetation and is usually used as forage grass for grazing animals in the Songnen grassland. In the mowing grassland, Leymus chinensis is harvested as hay, and farmers sell the hay to livestock farms. Alfalfa has high saline and alkaline tolerance, and it has been introduced into the Songnen grassland as a high forage plant due to the high N and protein content [22]. The planting of forage grass used in this study could improve the income of local farmers and the development of animal husbandry.

Experimental Design
This experiment was organized as a completed block design with five land use treatments. In early May 2011, four adjacent blocks (each 60 × 50 m, 2 m buffer between the blocks) in the study area based on similar land use history were identified. Before this experiment (2004-2010), farmers grew rain fed maize (Zea mays L.) and sunflower (Helianthus annuus) in these blocks, following the traditional planting practices in Northeast China, which consists of plowing the soil down to 20 cm depth and applying 50-96 kg N ha −1 , 20-45 kg P ha −1 , and 15-45 kg K ha −1 fertilizers into the soils. The soil properties in these four blocks were homogeneous due to the continuous plowing. The five land use treatments consisted of corn cropland (corn, used as an indication of how the revegetation influences the soils in this study), alfalfa perennial forage land (alfalfa), monoculture grassland of Leymus chinensis (MLG), monoculture grassland of Leymus chinensis for hay (Mowing) once a year (MLG + M), and successional regrowth grassland (SRG) (Figure 2). Leymus chinensis is the native vegetation and is usually used as forage grass for grazing animals in the Songnen grassland. In the mowing grassland, Leymus chinensis is harvested as hay, and farmers sell the hay to livestock farms. Alfalfa has high saline and alkaline tolerance, and it has been introduced into the Songnen grassland as a high forage plant due to the high N and protein content [22]. The planting of forage grass used in this study could improve the income of local farmers and the development of animal husbandry. In each block, two greater plots of 12 × 50 m were for corn and alfalfa treatments, while three plots of 6 × 50 m for land use treatments of MLG, MLG + M, and SRG. There was a 1 m buffer among the five plots. The was no irrigation under Land 2020, 9, 10 4 of 14 the five land uses in this study. More information about the treatments of land uses is presented in Figure 3 [22,26]. In each block, two greater plots of 12 × 50 m were for corn and alfalfa treatments, while three plots of 6 × 50 m for land use treatments of MLG, MLG + M, and SRG. There was a 1 m buffer among the five plots. The was no irrigation under the five land uses in this study. More information about the treatments of land uses is presented in Figure 3 [22,26].

Successional regrowth grassland (SRG)
The cropland was abandoned in 2011 in the SRG plots to restore grassland without any disturbance.
The dominant species in the SRG plots include Chloris virgate, Sonchus brachyotus, Chenopodium glaucum, etc. The aboveground (348 g m −2 ) and belowground (397 g m −2 , 0-20 cm depth) biomass was kept as litter to return to the soil. The aboveground biomass was mowed for hay once a year at the peak biomass.
The belowground (638 g m −2 , 0-20 cm depth) biomass was kept to return to the soils in these plots.

Monoculture grassland of Leymus chinensis (MLG)
Seeds of Leymus chinensis (Trin.) Tzvelev were sowed in May 2011 with a density of approximately 2000 seeds m −2 . Reseeding had a positive effect on the recovery of vegetation, and the aboveground biomass reached approximately 100-120 g m −2 in early September 2011. The aboveground (381 g m −2 ) and belowground (456 g m −2 , 0-20 cm depth) biomass was kept as litter to return to the soil.

Alfalfa perennial forage (Alfalfa)
Before 2014, the Alfalfa plots were no tillage cropland, and other practices were the same as those of the previously described corn cropland. However, the growth of corn was very poor in 2011 to 2013 due to the poor soil conditions and short term land use, and the no tillage cropland was changed to alfalfa forage land in May 2014 with a sowing density of approximately 1200 seeds m −2 . The aboveground (307 g m −2 ) and belowground (321 g m −2 , 0-20 cm depth) biomass were kept in these plots to increase soil fertility in 2014 and 2015.

Corn cropland (Corn)
Since 2011, the cropland has been under continuous corn monoculture. The Corn plots followed the traditional cropland practice in the Songnen grassland; this tradition consists of plowing the soil at least twice before the crop growing season down to 20 cm and fertilization (74 kg N ha −1 , 22 kg P ha −1 , and 41 kg K ha −1 ) twice per year at sowing and in mid-July. The corn straw was removed from the plots after harvest, while the corn root, stem base, and aerial root (which is 137 g m −2 ) were incorporated into soil during plowing.

Soil Sampling and Analysis
Soil sampling was performed using an auger (4 cm in diameter) in early September 2015. The sampling depth was 0-50 cm with an interval of 10 cm increments. Five randomly distributed subsamples from each plot were combined into a composite sample at each soil depth. After removing the visible vegetation materials and debris, soil samples were sieved through a 2 mm sieve, and then ground to pass through a 0.25 mm sieve for analyses.
Soil labile oxidizable carbon (LOC) was measured using the revised method defined by Chan et al. [27]. Available nitrogen (AN) was measured by the alkaline hydrolysis diffusion method [28]. The AN forms were primarily mixtures of ammonium nitrogen (NH3-N), nitrate nitrogen (NO3-N), and a small amount of water soluble organic nitrogen (e.g., amino acids and ammonium acyl, etc.). Available phosphorus (AP) was extracted with NaHCO3 at pH 8.5 and measured using UV spectrophotometer [28]. The AP forms were primarily the calcium phosphates due to the higher soil pH (Table 1) in the study area. Available potassium (AK) was measured based on the ammonium In each block, two greater plots of 12 × 50 m were for corn and alfalfa treatments, while three plots of 6 × 50 m for land use treatments of MLG, MLG + M, and SRG. There was a 1 m buffer among the five plots. The was no irrigation under the five land uses in this study. More information about the treatments of land uses is presented in Figure 3 [22,26].

Successional regrowth grassland (SRG)
The cropland was abandoned in 2011 in the SRG plots to restore grassland without any disturbance.
The dominant species in the SRG plots include Chloris virgate, Sonchus brachyotus, Chenopodium glaucum, etc. The aboveground (348 g m −2 ) and belowground (397 g m −2 , 0-20 cm depth) biomass was kept as litter to return to the soil.

Monoculture grassland of Leymus chinensis for hay (MLG + M)
Seeds of Leymus chinensis (Trin.) Tzvelev were sowed in May 2011 with a density of approximately 2000 seeds/m 2 . Reseeding had a positive effect in recovering the vegetation.
The aboveground biomass was mowed for hay once a year at the peak biomass.
The belowground (638 g m −2 , 0-20 cm depth) biomass was kept to return to the soils in these plots.

Monoculture grassland of Leymus chinensis (MLG)
Seeds of Leymus chinensis (Trin.) Tzvelev were sowed in May 2011 with a density of approximately 2000 seeds m −2 . Reseeding had a positive effect on the recovery of vegetation, and the aboveground biomass reached approximately 100-120 g m −2 in early September 2011. The aboveground (381 g m −2 ) and belowground (456 g m −2 , 0-20 cm depth) biomass was kept as litter to return to the soil.

Alfalfa perennial forage (Alfalfa)
Before 2014, the Alfalfa plots were no tillage cropland, and other practices were the same as those of the previously described corn cropland. However, the growth of corn was very poor in 2011 to 2013 due to the poor soil conditions and short term land use, and the no tillage cropland was changed to alfalfa forage land in May 2014 with a sowing density of approximately 1200 seeds m −2 . The aboveground (307 g m −2 ) and belowground (321 g m −2 , 0-20 cm depth) biomass were kept in these plots to increase soil fertility in 2014 and 2015.

Corn cropland (Corn)
Since 2011, the cropland has been under continuous corn monoculture. The Corn plots followed the traditional cropland practice in the Songnen grassland; this tradition consists of plowing the soil at least twice before the crop growing season down to 20 cm and fertilization (74 kg N ha −1 , 22 kg P ha −1 , and 41 kg K ha −1 ) twice per year at sowing and in mid-July. The corn straw was removed from the plots after harvest, while the corn root, stem base, and aerial root (which is 137 g m −2 ) were incorporated into soil during plowing.

Soil Sampling and Analysis
Soil sampling was performed using an auger (4 cm in diameter) in early September 2015. The sampling depth was 0-50 cm with an interval of 10 cm increments. Five randomly distributed subsamples from each plot were combined into a composite sample at each soil depth. After removing the visible vegetation materials and debris, soil samples were sieved through a 2 mm sieve, and then ground to pass through a 0.25 mm sieve for analyses.
Soil labile oxidizable carbon (LOC) was measured using the revised method defined by Chan et al. [27]. Available nitrogen (AN) was measured by the alkaline hydrolysis diffusion method [28]. The AN forms were primarily mixtures of ammonium nitrogen (NH3-N), nitrate nitrogen (NO3-N), and a small amount of water soluble organic nitrogen (e.g., amino acids and ammonium acyl, etc.). Available phosphorus (AP) was extracted with NaHCO3 at pH 8.5 and measured using UV spectrophotometer [28]. The AP forms were primarily the calcium phosphates due to the higher soil pH (Table 1) in the study area. Available potassium (AK) was measured based on the ammonium

Soil Sampling and Analysis
Soil sampling was performed using an auger (4 cm in diameter) in early September 2015. The sampling depth was 0-50 cm with an interval of 10 cm increments. Five randomly distributed sub-samples from each plot were combined into a composite sample at each soil depth. After removing the visible vegetation materials and debris, soil samples were sieved through a 2 mm sieve, and then ground to pass through a 0.25 mm sieve for analyses.
Soil labile oxidizable carbon (LOC) was measured using the revised method defined by Chan et al. [27]. Available nitrogen (AN) was measured by the alkaline hydrolysis diffusion method [28]. The AN forms were primarily mixtures of ammonium nitrogen (NH 3 -N), nitrate nitrogen (NO 3 -N), and a small amount of water soluble organic nitrogen (e.g., amino acids and ammonium acyl, etc.). Available phosphorus (AP) was extracted with NaHCO 3 at pH 8.5 and measured using UV spectrophotometer [28]. The AP forms were primarily the calcium phosphates due to the higher soil pH (Table 1) in the study area. Available potassium (AK) was measured based on the ammonium acetate extracted and emission flame spectrophotometer method [28]. The AK forms were primarily Land 2020, 9, 10 5 of 14 mixtures of exchangeable potassium and water soluble potassium. Furthermore, for the purpose of clearly understanding the nature of soils in the study area, a 1:5 soil:water solution was used to measure the soil pH and electrical conductivity (EC) using the PHS-3C instrument and the DDS-307 instrument, respectively.

Statistical Analysis
The soil stratification by certain properties (e.g., SOC, total N, total P, etc.) is very common, and the stratification ratio (SR) is widely used as a crucial indicator of soil condition [15]. A higher SR of soil properties indicates better soil conditions, because SR of degraded soils is usually less than 2 regardless of climatic or soil conditions [15]. The improvement of soil quality under specific land use is conducive to plant growth and agricultural sustainability [29,30]. Revegetation on the cropland will increase the input of organic matter and thus alter the SR of soil properties, which will provide an indication of soil responses to specific plant cover. The SR were calculated for each land use as follows: where AN t is the content of LOC, AN, AP, and AK in the 0-10 cm depth; AN s is the corresponding content of LOC, AN, AP, and AK in the 10-20 and 20-30 cm depth. A unitary soil available nutrient is not complete to reveal the changes within the soil environment because soil available nutrients do not always respond similarly to different management practices [22]. Therefore, the comprehensive assessment of the responses of a series of soil available nutrients and LOC to factors of change is required. However, the various responses of LOC and soil available nutrients to land use change might result in inaccurate conclusions on soil quality and thus limit the suitability of LOC and soil available nutrients as soil quality indicators. The geometric mean and sum scores are two general indices to combine the variables with diverse units and ranges into one variable, which could clearly indicate the actual influences of environmental factor changes on these variables [31]. Here, the geometric means of LOC and available nutrients (GMSN) under different land uses and soil depths are calculated as follows: where LOC, AN, AP, and AK are oxidizable labile C, available nitrogen, available phosphorus, and available potassium, respectively.
Land 2020, 9, 10 6 of 14 The simple sum of the series of soil available nutrients and LOC with different units and ranges of variation may cover up the changes in some soil nutrients. Therefore, data normalization is needed for all the measured soil available nutrients and LOC before the sum scores of LOC and available nutrients (SSAN) are calculated. The min-normalization is a well explored approach to convert the data with different units or variation ranges into a dimensionless pure value, so that the data can remove the unit limit and can be easily compared and weighted [26]. The SSAN under different land uses and soil depths is as follows: where S i is the score of LOC, AN, AP, and AK after data normalization; X is the measured value, and X min is the minimum value of each soil nutrient observed in this study; n is the number of soil nutrients. We used one way ANOVA to analyze the influences of land use types on the LOC and soil available nutrients, soil pH, EC, SR, GMSN, and SSAN. Mean differences of soil available nutrient contents, LOC, SR, GMSN, and SSAN among land use treatments were examined using the least significant difference test (LSD). All comparisons were considered significant if p < 0.05. The mean and standard error of each soil property measured were provided at each soil depth under a given land use treatment. All data analyses were performed with SPSS 16.0 for Windows (SPSS, Inc., Chicago, USA).

Changes in Soil pH and EC
Soil pH in the study area was notably high ( Table 1). The values of soil pH were all more than 9.00 at the 0-50 cm depth; especially at the 10-50 cm depths, the values were close to or more than 10.00. Soil pH was not affected by the land use conversions. The average values of soil pH at the 0-50 cm depth were 9.88, 9.82, 9.83, 9.76, and 9.68 for corn, alfalfa, MLG, MLG + M, and SRG treatment, respectively.
Similar to the soil pH, the EC values in the subsoil (10-50 cm) were higher than that at the surface soil (0-10 cm). The average values of EC in the 0-50 cm depth were 333, 317, 430, 399, and 376 µS cm -1 for corn, alfalfa, MLG, MLG + M, and SRG treatment, respectively (Table 1). There was no significant difference of EC among the land use treatments because of the narrow values of EC in the same soil depth.

Changes in LOC, AN, AP, and AK Content
The LOC content under the land use of SRG was remarkably higher than that under corn in the 0-10 cm depth, while it was significantly higher under alfalfa in the 10-20 cm depth than the corn and MLG + M treatment ( Figure 4A). In the 20-50 cm depth, the highest LOC content was found under the MLG treatment. The average LOC contents in the 0-50 cm depth were 32% (0.56 g kg −1 ), 28% (0.49 g kg −1 ), 15% (0.26 g kg −1 ), and 32% (0.57 g kg −1 ) higher under alfalfa, MLG, MLG + M, and SRG treatment, respectively, than that under corn treatment.
Land use conversions significantly (F = 8.76, p = 0.001) changed the AN contents ( Figure 4B). The highest AN content was found under corn treatment in the 0-50 cm depth. In addition, the AN contents under corn treatment in the 20-30 cm and 40-50 cm depth were significantly higher than those under all the revegetation land except the alfalfa treatment in the 20-30 cm depth. The average AN contents in the 0-50 cm depth were 15% (6.3 mg kg −1 ), 19% (8.0 mg kg −1 ), 34% (14.9 mg kg −1 ), and 27% (11.8 mg kg −1 ) lower under alfalfa, MLG, MLG + M, and SRG treatment, respectively, than under corn treatment.  kg −1 , respectively. The highest AK contents were all found under the MLG treatment in the 0-50 cm depth ( Figure 5B). However, significant differences were only found between MLG and corn treatment in the 20-40 cm depth and between MLG and alfalfa treatment in the 30-40 cm depth. The average AK contents in the 0-50 cm depth under corn, alfalfa, MLG, MLG + M, and SRG treatment were 102.9, 112.2, 137.5, 108.7, and 125.8 mg kg −1 , respectively.

Changes in SR, GMSN, and SSAN
The SR of LOC, AN, and AK in the 0-10/10-20 cm and in the 0-10/20-30 cm ( Figure 6) were not affected by the different land uses. Land uses of corn and alfalfa had remarkably higher SR values of

Discussion
The low content of soil organic carbon can limit microbial biomass and activity, nutrient cycling, soil structure formation, etc., and therefore indirectly limit plant growth [22]. Increasing the content of soil organic carbon and soil available nutrients is the common approach to improve soil productivity and agricultural sustainability. Land use changes could significantly alter the inputs and outputs of soil organic matter, thus resulting in the variations in the content and circulation of soil labile carbon and soil nutrients [32,33]. The present study showed that conversion of cropland to revegetation land increased the LOC content in the 0-50 cm depth ( Figure 4A). Moreover, the increase of LOC content mainly occurred in the surface soil. Compared with the corn treatment, the LOC contents under revegetation land were 33%, 33%, and 20% higher in the 0 to 10, 10 to 20, and 20 to 30 cm depths, respectively. However, there were no significant differences for LOC contents between corn and revegetation treatments in the 30 to 40 and 40 to 50 cm depths. The higher LOC contents under revegetation land in surface soil (0-30 cm) were probably associated with the accumulation of above-and below-ground biomass incorporated into the surface soils [34,35]. In addition, revegetation on the cropland could reduce the loss of LOC in fine soil fractions caused by rain and wind erosion, thus increasing the LOC content [22,33]. Soil texture can affect the soil aggregation processes and, therefore, influences the soil capacity to sequester organic carbon [36]. Tian et al. [37] in the alpine grassland on the Tibetan Plateau reported that soil organic carbon and total nitrogen

Discussion
The low content of soil organic carbon can limit microbial biomass and activity, nutrient cycling, soil structure formation, etc., and therefore indirectly limit plant growth [22]. Increasing the content of soil organic carbon and soil available nutrients is the common approach to improve soil productivity and agricultural sustainability. Land use changes could significantly alter the inputs and outputs of soil organic matter, thus resulting in the variations in the content and circulation of soil labile carbon and soil nutrients [32,33]. The present study showed that conversion of cropland to revegetation land increased the LOC content in the 0-50 cm depth ( Figure 4A). Moreover, the increase of LOC content mainly occurred in the surface soil. Compared with the corn treatment, the LOC contents under revegetation land were 33%, 33%, and 20% higher in the 0 to 10, 10 to 20, and 20 to 30 cm depths, respectively. However, there were no significant differences for LOC contents between corn and revegetation treatments in the 30 to 40 and 40 to 50 cm depths. The higher LOC contents under revegetation land in surface soil (0-30 cm) were probably associated with the accumulation of above-Land 2020, 9, 10 11 of 14 and below-ground biomass incorporated into the surface soils [34,35]. In addition, revegetation on the cropland could reduce the loss of LOC in fine soil fractions caused by rain and wind erosion, thus increasing the LOC content [22,33]. Soil texture can affect the soil aggregation processes and, therefore, influences the soil capacity to sequester organic carbon [36]. Tian et al. [37] in the alpine grassland on the Tibetan Plateau reported that soil organic carbon and total nitrogen stocks positively correlated with clay content and silt content, while they negatively related to sand content. Land use changes can indirectly affect soil texture through the redistribution of soil by erosional processes or tillage. Revegetation on the cropland in this study could reduce the soil erosion by increasing the vegetation cover and decreasing soil disturbance, thus indirectly affecting the content of LOC and soil available nutrients.
Compared with corn treatment, revegetation did not increase the AN contents in the study area, and the corn treatment had the highest AN content in the 0 to 50 cm depth ( Figure 4B). This might be due to the fertilization management in corn treatment, which applied approximately 74 kg N ha −1 every year. Another reason for the higher AN contents under corn treatment could partially result from the short term revegetation under the revegetation land, which had limited effects on the accumulation of AN and other soil nutrients. Besides, no significant differences among the forage and grasslands also suggested negligible effects of short term revegetation on the AN contents in the study area. The higher AN contents under the corn treatment were similar to the results by Zhang et al. [38] in Guizhou, China, who also reported that the AN content under fertilized and plowed cropland was higher than that under grassland and forestland. Soil AP and AK contents were not significantly different under most land use treatments ( Figure 5), indicating that the short term land use treatments did not change the AP and AK contents in northeastern China. Similar to the changes in AN content, the negligible effects of short term revegetation on soil AP and AK may be the primary reason for the narrow changes in AP and AK content under the five land use treatments. These results were in agreement with the findings of Zhao et al. [39] in another region of Songnen plain, who also found that the changes in AP and AK content under cropland and grassland were very limited.
The SR of soil parameters was used as an indicator of the dynamics soil quality, and it could detect the management induced changes in the soil profiles of agricultural systems [15]. The increase in SR values of LOC and soil nutrient indicated the improvement of soil quality due to the accumulation of LOC and soil nutrients in the surface soil [30,40]. Land use treatments had no significant effects on the SR values of LOC, AN, AP, and AK contents at depths of 0-10/10-20 cm and 0-10/20-30 cm except the SR of AP at the depth of 0-10/20-30 cm, suggesting that short term revegetation had limited effects on the soil available nutrients in northeastern China. Studies in Columbia and Georgia showed that the SR values of SOC and total nitrogen were >2 under no tillage management, indicting an improvement of soil quality [29]. Peregrina et al. [41], Corral-Fernandez et al. [42], Francaviglia et al. [40], and Deng et al. [15] confirmed this finding, arguing that a high SR value (usually >2) indicated a better soil quality and contribution to agriculture sustainability. Our results showed that the SR values of LOC and soil available nutrients at the depth of 0-10/10-20 cm were mostly <2, and the SR values at the depth of 0-10/20-30 cm were mostly >2, indicating that soils under the same land use treatments had different soil quality. Similarly, the study by Deng et al. [15] also found that the SR values at the depth of 0-20/20-40 cm were generally higher than those at the depth of 0-5/10-20 cm found by Wang et al. [43] in the same region of the Loess Plateau. The SR values of LOC and soil nutrients in different soil depths in response to land use treatment were not consistent, suggesting that standard SR values of soil properties are needed in future studies to make the comparisons of soil quality under different management practices and different regions easier. Therefore, the SR values at the depth of 0-10/10-20 cm may be well suitable as a standard for evaluating significant changes in surface soils induced by management practices.
In this study, three soil available nutrients including AN, AP, and AK contents and LOC were evaluated, but similar trends were not found (Figures 4 and 5). In fact, it is difficult to draw meaningful conclusions about soil quality changes when univariate indicators are used to analyze datasets involving many soil properties and reveal the changes within the soil environment [44]. The two indices of GMSN and SSAN were able to overcome the above weaknesses, and they were used as useful indictors of soil quality in other studies [22,26,31]. However, the results in this study showed no significant differences of GMSN and SSAN among the five land use treatments at each soil depth except SSAN under the MLG + M treatment in the 0 to 10 cm depth (Figure 7), indicating that short term conversions of cropland to revegetation land had limited demonstrable influences on the soil available nutrients and LOC in the salt affected region of Songnen plain. The inconclusive results suggested that a long term study is needed to examine the responses of the LOC and soil available nutrients to long term revegetation in northeastern China.

Conclusions
The present results showed that revegetation on the cropland enhanced the LOC contents and decreased the AN contents in the 0-50 cm depth compared with the Corn treatment, and the changes in AP and AK contents were very limited after the land use conversions. The SR values in different soil depths in response to land uses were not consistent, suggesting that standard SR values of soil properties are needed in the future studies and that the SR values at the depth of 0-10:10-20 may be suitable as the standard considering the notable changes in surface soils induced by management practices. However, more studies are needed to examine if the SR value at the 0-10:10-20 cm is suitable in other managements or regions. The values of SR, GMSA, and SSAN were not affected by the land use changes, indicating short term revegetation on the cropland had limited influences on the changes in soil nutrients and LOC in northeastern China. Compared with AG treatment, values of GMSA and SSAN were slightly lower than other land use treatments. These results were mainly due to the very short term (five years) revegetation because revegetation may need more time to be incorporated. Therefore, more studies are needed to assess the long term (more than 10 years) effects of revegetation on soil properties in the Songnen grassland in the future. Although changes in soil available nutrients were given in this study, variations in soil microbial populations, which are more sensitive to changes in land uses than soil nutrients, were not mentioned. The influences of short term revegetation on soil quality need to be comprehensively assessed. In addition, we recommend that farmers in Northeast China should use revegetation to rehab grassland in areas with poor quality soils in the long run.