Interplay between Selected Chemical and Biochemical Soil Properties in the Humus Horizons of Grassland Soils with Low Water Table Depth

: Grasslands are the most carbon-rich of all agricultural ecosystems, but are also the most endangered. The global area of grassland decreased during the 20th century, mainly due to conversion to arable land, improper management, and abandonment. Due to ongoing climate change, maintenance of an adequate level of soil organic matter is of primary importance, not only to maintain the productive function of the soils, but also to ensure their role as carbon stores. The main aim of this study was to assess the linkages between selected chemical and biochemical soil properties in alluvial grassland soils, characterized by a low water table. The area under study was located in the Koło Basin, central Poland. Soil parameters, such as total organic carbon, total nitrogen, pH, phosphorus, magnesium, and potassium contents, as well as enzymatic activity and soil microbial structure were studied. Positive correlations were observed between total organic carbon content and the following enzymatic activities: dehydrogenase (r = 0.63), acid phosphatase (r = 0.69), and alkaline phosphatase (r = 0.86). There was a signiﬁcant correlation between fungi abundance and phosphorus and potassium contents, and between actinobacteria abundance and total organic carbon content


Introduction
Grasslands are the most carbon-rich of all agricultural ecosystems, storing ca.34% of the total global carbon stock held in terrestrial ecosystems [1].Moreover, permanent or semi-natural grasslands play an important role in maintaining biodiversity, while also contributing to agricultural production through livestock grazing or fodder production [2].However, the global grassland area has declined during the 20th century, mainly due to conversion to arable land, improper management, and abandonment [3].In Europe, grasslands are currently considered as among the most threatened ecosystems [4].For instance, the area of grassland in Poland has markedly declined in recent decades, and now constitutes only ca.20% of total agricultural land [2].In central Poland, grassland degradation has been especially observed in the vicinity of open-pit mines [5,6].Open-pit mining leads to detrimental changes in local hydrological conditions (depression cone formation) that might directly affect adjacent ecosystems and impact the processes in the surface soil layers in particular [7,8].
Organic matter plays an important role in enhancing soil fertility as it affects the physical, chemical, and biological properties of the soil [5].Due to ongoing climate change, maintenance of an adequate level of organic matter is of primary importance not only to maintain the productive function of the soil, but also to ensure the role of soils as carbon stores [6].The activity of soil microorganisms has been identified as a major factor that influences the soil carbon storage potential [9,10].Moreover, the abundancy and activity of microorganisms determine the direction and nature of biochemical processes in the soil, thereby directly affecting soil fertility [11].Factors that influence soil microbial structure and activity include soil temperature, changes in soil hydrology/moisture content, nutrient availability, and soil pH [12,13].Enzyme activity is conditioned by the presence of organic matter.Accordingly, microbiological and biochemical indicators can be used to predict the changes caused by anthropogenic activities, and are considered highly sensitive indicators of the state of the soil environment [14,15].
Sustainable management of grasslands is an important issue, especially in the context of biodiversity protection and carbon sequestration [6].This statement is in line with the European Green Deal, in which the crucial importance of soil fertility protection and increasing soil organic matter levels are highlighted.However, before sustainable management practice can be implemented, detailed knowledge of the properties and main determinants for the processes that occur in the surface soil layers is required, as these layers constitute a critical environment for plant roots and microorganisms [16].
The main aim of this study was to assess the linkages between selected chemical and biochemical soil properties in alluvial grassland soils, characterized by a permanently low water table.A greater understanding of these dependencies is of crucial importance, as climate change and human activities are likely to limit water access in grasslands globally.We address the following research hypotheses: (1) soil pH is the most critical factor among chemical soil properties, as it influences the structure of microbial communities in soils with permanently low water tables, and (2) enzyme activity is mainly determined by total organic carbon and nitrogen contents in the soil.

Study Area
The study area is located in the Koło Basin (southern Wielkopolska Lowland, central Poland).All 10 sampling sites were located on intensively used grasslands.According to the Detailed Geological Map of Poland [17], all the soils under study were developed from alluvial materials, sands, and silts.All the sampling sites were established within the contours of alluvial soils (F) according to a soil-agricultural map.According to the Köppen-Geiger climate classification, this region of Poland is located in the fully humid warm temperate climate zone with warm summers [18].The mean annual air temperature (MAAT) and the mean annual precipitation (MAP) in this region are 8.7 • C and 520 mm, respectively [19].However, in recent years, the air temperature has increased slightly (+1.0 • C), whereas the mean annual rainfall has decreased (−25 mm) [20].
as Fluvic Phaeozems (P1, P3, P4, P5, P7, P8, P9), Fluvic Cambisols (P2, P6), and Umbric Gleysol (P10).Soil samples for laboratory analyses were collected from the humus horizon in each soil profile from three points within one sampling site in order to integrate potential variability in the soil.Sampling sites were georeferenced in the field with a Garmin GPSMap 60 CSx receiver with an accuracy of 3 m (Table 1).The samples were collected in plastic bags and placed in a lightproof insulated box that contained ice packs to ensure constant low temperature during transportation.Water table depth during the sampling campaigns ranged from 1-3 m below ground level (b.g.l), with the deepest positions observed in sites P5 (2.7 m b.g.l) and P10 (3.0 m b.g.l) (Table 1).In the study area, a very low water table depth (<5 m b.g.l) has also been recorded during long-term soil monitoring; the potential effect of the open-pit Drzewce lignite mining operation [8,24].

Chemical Analyses
In the laboratory, the soil samples were air-dried, disaggregated, homogenized, and sieved through a 2 mm sieve.The following parameters were measured in dry samples: total organic carbon (TOC), total nitrogen (TN), and total sulfur (TS) content with a Vario-  The samples were collected in plastic bags and placed in a lightproof insulated box that contained ice packs to ensure constant low temperature during transportation.Water table depth during the sampling campaigns ranged from 1-3 m below ground level (b.g.l), with the deepest positions observed in sites P5 (2.7 m b.g.l) and P10 (3.0 m b.g.l) (Table 1).In the study area, a very low water table depth (<5 m b.g.l) has also been recorded during long-term soil monitoring; the potential effect of the open-pit Drzewce lignite mining operation [8,24].

Chemical Analyses
In the laboratory, the soil samples were air-dried, disaggregated, homogenized, and sieved through a 2 mm sieve.The following parameters were measured in dry samples: total organic carbon (TOC), total nitrogen (TN), and total sulfur (TS) content with a Vario-Max CNS analyzer; calcium carbonate content with the Scheibler volumetric method [25]; soil pH potentiometrically in a suspension of distilled water at a ratio of 1:2.5; plant available phosphorus (P) and potassium (K) contents with the Egner-Riehm method (extracted with Ca lactate at pH 3.6); plant available magnesium (Mg) content with the Schachtschabel method (extracted with 0.0125 M CaCl 2 ).The K and Mg content was measured with the AAS method using a Varian SpectrAA 220FS spectrometer.The P content in the extracts was measured with the colorimetric method using an Agilent Cary 60 UV-Vis spectrophotometer, with molybdate and SnCl 2 as reactants.

Biochemical Analyses
Soil microorganism count was measured with the plate method on adequate selective substrates with 5 replicates.The total count of colony forming units (CFU) of heterotrophic bacteria, actinobacteria, fungi and Azotobacter bacteria was measured.Heterotrophic bacteria count was measured on ready-made Merck standard agar medium (28 • C for 5 days).Fungi count was measured on a Martin substrate [26] after a 5-day incubation at 24 • C. Actinobacteria count was measured on a selective Pochon substrate [27], where the plates were incubated for 7 days at 26 • C. Azotobacter bacteria count was determined on a selective medium according to the method developed by [28], after 5 days of incubation at 24 • C.
Dehydrogenase activity (DhA) was measured with the Thalmann method [29], using a 1% solution of triphenyl tetrazolium chloride (TTC) as a substrate.DhA was expressed as mg TPF kg −1 24 h −1 .Acid phosphatase activity (PhacA) and alkaline phosphatase activity (PhalA) were measured according to the method developed by Tabatabai and Bremner [29], using a 0.8% sodium p-nitrophenyl phosphate solution as substrate.The buffer pH values were 5.4 and 8.5, respectively.PhacA and PhalA were expressed as mmol PNP kg −1 h −1 .Urease activity (UA) was measured with the method developed by [29], using a 2.5% urea solution as substrate.Activity was expressed as mg N-NH 4 + kg −1 h −1 .All the measurements were made in 3 replicates.

Data Processing
Based on chemical soil properties similarities (confirmed by hierarchical and nonhierarchical clustering, Figure 2), the individual soils were assigned to the following groups: Group I (P2, P9), Group II (P3, P7, P8, P10), and Group III (P1, P4, P5, P6).Hereafter, all results reported in this paper will correspond to these groups.The soil properties values were expressed as means and compared between groups by Principal Component Analysis (PCA) [30].We also applied PCA to test our hypotheses with regard to the relationships between microorganism structure, biochemical activity, and chemical soil properties.The detailed correlation between microorganism structure, biochemical properties, and the other soil properties was analyzed by Pearson correlations.For all calculations, we used the statistical package Statistica 13 (StatSoft Inc., Tulsa, OK, USA) and Canoco v. 5.0 [30].

Chemical Soil Properties
Mean TOC content for the individual sampling plots in the surface humus horizons ranged from 13.1 to 53.4 g kg −1 , with the greatest content recorded in Group II soils (Table

Chemical Soil Properties
Mean TOC content for the individual sampling plots in the surface humus horizons ranged from 13.1 to 53.4 g kg −1 , with the greatest content recorded in Group II soils (Table 2).Similarly, the greatest mean TN (5.23 g kg −1 ) and TS values (0.64 g kg −1 ) were also observed in Group II soils, and the lowest (1.32 and 0.18 g kg −1 , respectively) were observed in Group I soils (Table 2). 1 total organic carbon, 2 total nitrogen, 3 total sulfur, 4 total organic carbon to total nitrogen ratio, 5 total organic carbon to total sulfur ratio, 6 plant available magnesium, 7 plant available potassium, 8 plant available phosphorus.
The calculated ratio of TOC/TN and TOC/TS ranged between 9.8-10.2and 78.1-89.7,respectively.The observed pH values were characteristic of slightly acidic soils (6.3-Group II) and acidic soils (5.8-GroupsI and III).For plant available forms of P, K, and Mg, the humus horizons were characterized by very low P content (19.8-45.0g kg −1 ) and low (73.4 g kg −1 ) to very high (267 g kg −1 ) plant available K content (Table 2).The studied soils were characterized by low (29.3 g kg −1 ) and very high (116 g kg −1 ) Mg content.In general, the highest plant available P and K contents were observed in Group I, while the highest Mg content was recorded in Group II samples.The main differences between groups are summarized in Figure 3.

Microbial Community and Enzymatic Activity
In our study, heterotrophic bacteria were the most numerous group of microorganisms, followed by actinobacteria and fungi (Table 3).Overall, the largest number of heterotrophic bacteria was recorded in Group III soils, while the lowest counts were observed in Group I soils.The largest number of actinobacteria and Azotobacter were recorded in Group II soils, followed by Group III soils, with the lowest counts observed in Group I soils.In contrast, fungi numbers were largest in Group I soils, lower in Group II soils, and lowest in Group III soils.In summary, heterotrophic bacteria, actinobacteria, and Azotobacter were found to be dominant in Groups II and III, while fungi were found to be dominant in Group I soils.
The greatest DhA value was recorded in the humus horizons of Group II soils, a slightly lower value was found in Group I samples, and the lowest value was observed in Group III samples (range: 3.50-5.95mg TPF kg −1 dm soil 24 h −1 ).The greatest UA was found in Group II samples, while slightly lower values were observed in Group I and III samples (range: 36.9-46.9mg N-NH 4 + kg −1 dm soil h −1 ).In the analyzed soils, PhacA and PhalA were greatest in Group II soil samples, lower values were recorded in Group III soil samples, and the lowest values was found in Group I samples (Table 3).The greatest level of enzyme activity was recorded in Group II, while Groups I and III were lowest.Analysis.Abbreviations: I-III denote groups according to prior classification, P1-P10 denote sampling sites, TOC-total organic carbon, TN-total nitrogen, TS-total sulfur, TOC/TN-total organic carbon to total nitrogen ratio, TOC/TS-total organic carbon to total sulfur ratio, Mg-plant available magnesium, K-plant available potassium, P-plant available phosphorus, pH-soil pH.

Microbial Community and Enzymatic Activity
In our study, heterotrophic bacteria were the most numerous group of microorganisms, followed by actinobacteria and fungi (Table 3).Overall, the largest number of heterotrophic bacteria was recorded in Group III soils, while the lowest counts were observed in Group I soils.The largest number of actinobacteria and Azotobacter were recorded in Group II soils, followed by Group III soils, with the lowest counts observed in Group I soils.In contrast, fungi numbers were largest in Group I soils, lower in Group II soils, and lowest in Group III soils.In summary, heterotrophic bacteria, actinobacteria, and Azotobacter were found to be dominant in Groups II and III, while fungi were found to be dominant in Group I soils.

Figure 3. Differences in soil properties between groups of sites described by Principal Component
Analysis.Abbreviations: I-III denote groups according to prior classification, P1-P10 denote sampling sites, TOC-total organic carbon, TN-total nitrogen, TS-total sulfur, TOC/TN-total organic carbon to total nitrogen ratio, TOC/TS-total organic carbon to total sulfur ratio, Mg-plant available magnesium, K-plant available potassium, P-plant available phosphorus, pH-soil pH. 1 dehydrogenases activity, 2 urease activity, 3 acid phosphatase activity, 4 alkaline phosphatase activity.

Linkages between Chemical and Biochemical Soil Properties
Our PCA analysis showed that, in general, the first ordination axis explained ca.67% of the variability between samples, and the second axis explained ca.20% (Figure 4).The first axis is related to the differences between Group I and the other two groups.The second axis explains the differences between Groups II and III.The greatest enzymatic activity was detected in Group II soil samples and was related to the large number of actinobacteria and Azotobacter bacteria, which prefer soils with high TN and TOC contents and relatively high pH levels.The presence of fungi and heterotrophic bacteria was related to the relatively high K and P contents in the Group I soil samples.Detailed analysis by Pearson correlations demonstrated that these relationships were significant (p < 0.05) (Figure 5).The correlation between fungi abundance and P and K content was very high (r = 0.78 and r = 0.86 respectively).Moreover, the correlation between actinobacteria abundance and TOC/TS was high (r = 0.64).However, we did not find a significant relationship between the different groups of microorganisms and observed enzymatic activities.

Discussion
The results of our study indicate a significant positive correlation between fungi numbers and P and K contents.A similar effect of P content on fungi abundance was reported by Graça et al. [31] for grassland soils in Ireland.Moreover, Chmolowska et al. [32], who studied grassland soils in the Polish Western Outer Carpathians, reported that the availability of P and K significantly affects root colonization by arbuscular mycorrhizal fungi.According to Glaser et al. [33], fungi are capable of utilizing complex and difficultto-access forms of organic matter, unlike bacteria, which use readily available substrates.
In our study, soil pH did not correlate with any of the analyzed biochemical parameters.Similar findings were also reported by Furtak et al. [34], who studied Fluvisols from the Vistula floodplains in Poland.In contrast, the correlation between microbiological properties and enzymatic activity and pH has been reported to be very common [35,36].A number of recent studies have confirmed that pH is the main factor that influences the structure of the microbial community in the surface horizons of grasslands soils [37].Based on our results here, the hypothesis that soil pH is one of the main factors influencing the structure of microbial communities must be rejected.
We found a positive correlation between actinobacteria abundance and TOC/TS.Microorganisms play a key role in both the carbon and sulfur cycles in the soil, right next to temperature or soil pH [38][39][40].It should be noted that the TOC/TS ratio is less stable than, for example, the TN/TS ratio.This affects mineralization and immobilization of Pearson correlation matrix between biochemical and chemical soil properties.Only significant correlations are marked with p < 0.05.Abbreviations: I-III denote groups according to prior classification, P1-P10-sampling sites, TOC-total organic carbon, TN-total nitrogen, TS-total sulfur, TOC/TN-total organic carbon to total nitrogen ratio, TOC/TS-total organic carbon to total sulfur ratio, Mg-plant available magnesium, K-plant available potassium, P-plant available phosphorus, pH-soil pH, DhA-dehydrogenase activity, PhacA-acid phosphatase activity, PhalA-alkaline phosphatase activity, UA-urease activity, HB-heterotrophic bacteria, AB-actinobacteria, F-fungi, AZ-Azotobacter.

Discussion
The results of our study indicate a significant positive correlation between fungi numbers and P and K contents.A similar effect of P content on fungi abundance was reported by Graça et al. [31] for grassland soils in Ireland.Moreover, Chmolowska et al. [32], who studied grassland soils in the Polish Western Outer Carpathians, reported that the availability of P and K significantly affects root colonization by arbuscular mycorrhizal fungi.According to Glaser et al. [33], fungi are capable of utilizing complex and difficultto-access forms of organic matter, unlike bacteria, which use readily available substrates.
In our study, soil pH did not correlate with any of the analyzed biochemical parameters.Similar findings were also reported by Furtak et al. [34], who studied Fluvisols from the Vistula floodplains in Poland.In contrast, the correlation between microbiological properties and enzymatic activity and pH has been reported to be very common [35,36].A number of recent studies have confirmed that pH is the main factor that influences the structure of the microbial community in the surface horizons of grasslands soils [37].Based on our results here, the hypothesis that soil pH is one of the main factors influencing the structure of microbial communities must be rejected.
We found a positive correlation between actinobacteria abundance and TOC/TS.Microorganisms play a key role in both the carbon and sulfur cycles in the soil, right next to temperature or soil pH [38][39][40].It should be noted that the TOC/TS ratio is less stable than, for example, the TN/TS ratio.This affects mineralization and immobilization of organic matter.In the absence of available sulfur, microorganisms, including actinobacteria, uptake the necessary amount of it from the soil, further immobilizing sulfur within the soil.In our study, the TOC/TS ratio was between 78.1-89.7,which indicates intense mineralization of soil organic matter.According to Korb and Jacobsen [41], sulfur is immobilized in the soil at TOC/TS ratios > 400:1, and is mineralized at TOC/TS ratios < 200:1.
All our study soils were characterized by elevated PhacA and low DhA.For the latter, this results from their low resistance to environmental stresses, such as soil water deficiency, which in turn proves the usefulness of these enzymes as indicators of soil biological health [42].For DhA, PhacA, and PhalA, we noted a positive correlation with TOC.In contrast, there was no correlation between UA and soil organic matter components.Our findings are similar to these reported by Krzy żniak and Lemanowicz [43], who assessed the enzymatic activity of grassland soils in the Kujawy region (north-central Poland) against the chemical properties of the soils and found similar relationships.We found a strong correlation between TOC and TN contents and PhacA and PhalA.The increase in available nitrogen content promotes PhalA [44].In addition, the level of dehydrogenase and phosphatase activities in soils is mainly determined by the TOC content.These findings fully confirm our second research hypothesis that enzyme activity is mainly determined by TOC and TN content in the soil.Usually, the activity of soil phosphatases is inversely proportional to the content of available P [45,46].In the case of our research, a negative correlation between PhalA and P was also noted.Moreover, PhalA was positively correlated with Mg content.Even low Mg concentrations in the soil can lead to an increase in the population of microorganisms, thereby stimulating PhacA and PhalA.These findings show that Mg content could be considered as an enzyme activator as it affects the rate of phosphatase-related reactions [47].

Conclusions
Our results showed that the analyzed biochemical parameters (enzymatic activity and microbial abundance) were mainly determined by organic matter components (TOC, TN, P, and Mg).The grassland sites in our study are characterized by permanent low water tables, thus, the surface soil horizons were often exposed to prolonged drying.This leads to increased mineralization of organic matter and decreased content.Thus, a reduction in energy resources (for microorganisms) and nutrient availability (for plants) will affect soil fertility levels within these ecosystems.Based on our results, it can be concluded that it is necessary to raise the groundwater table and improve grassland management to slow down organic matter mineralization.In the absence of these treatments, further deterioration of the biological activity of the studied soils will occur due to the lack of organic compounds in the soil.Good practice and prudent grassland management is essential to minimize the negative effects of climate change manifested in the form of prolonged droughts.Therefore, to be in line with the goals of the Paris Climate Agreement, it is critical to monitor carbon-rich ecosystems, as soil organic matter dynamics in agricultural areas are of crucial importance.

Figure 1 .
Figure 1.Location of (A) the study area and (B) sampling sites.

Figure 1 .
Figure 1.Location of (A) the study area and (B) sampling sites.

12 Figure 2 .
Figure 2. Results of classification: (a) groups by hierarchical agglomerative cluster analysis with Bray-Curtis similarity measure, and (b) groups by non-hierarchical k-means clustering.Abbreviations: P1-P10 denote sampling sites.

Figure 2 .
Figure 2. Results of classification: (a) groups by hierarchical agglomerative cluster analysis with Bray-Curtis similarity measure, and (b) groups by non-hierarchical k-means clustering.Abbreviations: P1-P10 denote sampling sites.

Figure 3 .
Figure3.Differences in soil properties between groups of sites described by Principal Component Analysis.Abbreviations: I-III denote groups according to prior classification, P1-P10 denote sampling sites, TOC-total organic carbon, TN-total nitrogen, TS-total sulfur, TOC/TN-total organic carbon to total nitrogen ratio, TOC/TS-total organic carbon to total sulfur ratio, Mg-plant available magnesium, K-plant available potassium, P-plant available phosphorus, pH-soil pH.

Figure 4 .
Figure 4. Differences in chemical and biochemical properties between soil groups based on Principal Component Analysis.Abbreviations: I-III denote groups according to prior classification, P1-P10denotes sampling sites, TOC-total organic carbon, TN-total nitrogen, TS-total sulfur, TOC/TN-total organic carbon to total nitrogen ratio, TOC/TS-total organic carbon to total sulfur ratio, Mg-plant available magnesium, K-plant available potassium, P-plant available phosphorus, pH-soil pH, DhA-dehydrogenase activity, PhacA-acid phosphatase activity, PhalA-alkaline phosphatase activity, UA-urease activity, HB-heterotrophic bacteria, ABactinobacteria, F-fungi, AZ-Azotobacter.

Figure 4 .
Figure 4. Differences in chemical and biochemical properties between soil groups based on Principal Component Analysis.Abbreviations: I-III denote groups according to prior classification, P1-P10 denotes sampling sites, TOC-total organic carbon, TN-total nitrogen, TS-total sulfur, TOC/TN-total organic carbon to total nitrogen ratio, TOC/TS-total organic carbon to total sulfur ratio, Mg-plant available magnesium, K-plant available potassium, P-plant available phosphorus, pH-soil pH, DhA-dehydrogenase activity, PhacA-acid phosphatase activity, PhalA-alkaline phosphatase activity, UA-urease activity, HB-heterotrophic bacteria, AB-actinobacteria, F-fungi, AZ-Azotobacter.

Table 1 .
Main characteristics of the sampling sites.

Table 1 .
Main characteristics of the sampling sites.