Measurements versus Estimates of Soil Subsidence and Mineralization Rates at Peatland over 50 Years (1966–2016)

Abstract

The size of peat subsidence at Solec peatland (Poland) over 50 years was determined. The field values for subsidence and mineralization were compared with estimates using 20 equations. The subsidence values derived from equations and field measurements were compared to rank the equations. The equations that include a temporal factor (time) were used to forecast subsidence (for the 20, 30 and 40 years after 2016) assuming stable climate conditions and water regime. The annual rate of subsidence ranged from 0.08 to 2.2 cm year⁻¹ (average 1.02 cm year ⁻¹). Equation proposed by Jurczuk produced the closest-matching figure (1.03 cm year⁻¹). Applying the same equation to calculate future trends indicates that the rate of soil subsidence will slow down by about 20% to 0.82 cm year⁻¹ in 2056. With the measured peat subsidence rate, the groundwater level (57–72 cm) was estimated and fed into equations to determine the contribution of chemical processes to the total size of subsidence. The applied equations produced identical results, attributing 46% of peat subsidence to chemical (organic matter mineralization) processes and 54%—to physical processes (shrinkage, organic matter consolidation). The belowground changes in soil in relation to groundwater level have been neglected lately, with GHGs emissions being the main focus.

1. Introduction

Peatlands are negatively impacted by human activity in a number of ways. The most important being: agricultural drainage (mainly for meadows and pastures, but also arable fields and plantations, including those for growing energy crops) [1,2,3,4,5,6,7,8,9]. For peat soils to be used for agriculture, groundwater levels must be lowered (40–80 cm) by way of land drainage. Large swaths of peatland in Europe, as well as in other regions of the world, have been drained. Particularly intense drainage efforts were conducted in Europe in the 19th century and in the first half of the 20th century. In Poland, more than 75% of low peat areas were drained for meadow and pasture use, as well as—to a lesser extent—arable land. Fens (lowland peats) are the predominant type of peatland in Poland, and their high fertility potential has made them an attractive target for conversion (into meadows and grazing land) by farmers for several centuries [1].

Drainage induces a sequence of extensive changes in the soil body. The first phase, which usually lasts from around five to fifteen (even twenty) years, involves shrinkage and consolidation (loss of volume) spurred by buoyancy and compaction, resulting in rapid surface lowering [10,11,12,13,14,15,16,17,18,19,20]. In the following years, the microbiological decomposition of plant litter becomes the predominant mechanism, leading to the formation of humus and the release of CO₂ into the atmosphere [21]. Repeated drying and rewetting causes the humus to coagulate, which forms hard, granular aggregates with low internal porosity, and thus lower plant-available water capacity [22,23]. Only a portion of the carbon contained in organic compounds is transformed into humus during microbiological decomposition of peat—most of it is oxidized into CO₂ [24]. The CO₂ generated by peat decomposition persists in the atmosphere, increasing net emissions and causing global warming [25,26]. On the other hand, some LMW (low-molecular-weight) organic compounds penetrate into the ground and surface waters as dissolved organic carbon (DOC), negatively affecting water quality [27,28,29,30].

Land subsidence is a function of two processes—changes in the geometry of void space in the soil [31], and reduction in organic material (by volume and mass). These two factors cumulatively cause land to subside. Subsidence is usually defined as a result of three drivers: shrinkage, consolidation, and peat oxidation. Oxidation of organic matter takes place in the aeration zone. This process is biochemical and lasts until the available organic matter (peat) is exhausted. In the long term, peat oxidation is the key factor in land subsidence and the resultant removal of organic peat soils (which are subsequently reclassified taxonomically to mineral soil) [7,18,22]. The process of converting peat soil into mineral post-peat soil is referred to as “peatland disappearance” [17].

Subsidence is divided into two main types: primary and secondary [32,33]. Primary subsidence occurs immediately after drainage, as the soil body sinks (subsides) under its own weight due to loss of buoyant force. This process happens relatively quickly and is virtually complete within 4 to 10 (20) years of drainage. Secondary subsidence is caused by several processes, the most influential being microbial decomposition (oxidation) of organic matter, shrinkage on drying, and leaching of soluble compounds. This phase can last for tens or even hundreds of years, depending on the thickness of the peat layer, the climate, the extent of drainage, and intensity of land use [11,12,20,34,35,36,37,38]. Organic soils subject to continuous drainage are not sustainable, and have limited life expectancies, which can be reliably determined on the basis of the subsidence rates of similar deposits [32]. Subsidence can also be exacerbated by disaster events, such as fires of peat deposits or deflation (erosion of loose humus particles by wind from organic soils used as arable land) [32,37,38,39].

The rate of subsidence is a function of multiple factors, including: peat type, degree of peat decomposition, mineral matter content (ash content), bulk density, initial thickness of peat deposit, drainage depth, climate, land use, nutrient content and fertilization applied, and duration of drainage [17,19,33,34,38,39,40,41,42,43,44]. It is difficult to take into account all of the above-mentioned factors in the field research. Therefore, a number of empirical equations (most often linear relationships) have been developed between the rate of peat subsidence and selected (relatively easily measurable) parameters, such as: the initial thickness of the peat deposit, the time that has elapsed since peatland drainage and the intensity of drainage (depth of drainage). For many equations the values of, inter alia, groundwater level and soil bulk density are required. A review of such equations was presented, among others, by Ilnicki [45], Jurczuk [13], Grzywna [17], Evans et al. [5], Oleszczuk et al. [19] and Ikkala et al. [20]. The average annual peat subsidence rate varies and ranges from 10 cm year⁻¹ in the first phase of drainage to approx. 1–3 cm year⁻¹ in the second phase.

Soil subsidence and CO₂ emissions should be considered to be economically detrimental, as it leads to deformation of the drainage system, irregular hydration across the field, and recurrent paludification (reswamping) [20]. In order to maintain usability for agriculture, drainage ditches need to be constantly deepened, which in turn leads to another cycle of land subsidence [38,39,46,47]. Emissions of GHGs (CO₂ and N₂O) from drained peats used for agriculture are a significant contributor to net greenhouse gas levels in the atmosphere. Total emission from drained organic soils was estimated to be nearly one billion tons CO₂eq annually, with CO₂ accounting for approx. 780 million tons—more than one-fourth of total net CO₂ emissions from agriculture, forestry, and land use [25].

Peatland subsidence negatively affects the agricultural use of organic soils, reduces their thickness and area, which in turn may even lead to their complete disappearance [38]. Therefore, it can be regarded as a synthetic measure of the degradation of drained peat soils. Knowledge on existing (past) subsidence and estimates of future subsidence at specific site conditions and dewatering intensities is essential to farmers, as well as to administrators, planners and land managers, so that they may make proper use of soils and protect them against degradation [47,48,49,50].

The main objective of the study was to compare the calculated and field-measured magnitude/rate of subsidence of the surface of drained peat soil in central Europe (Poland), which also made it possible to verify empirical equations for the magnitude of subsidence. On the basis of the field measurements and the verification of the formulas (with a time component), the predicted magnitudes of subsidence of these soils in the long term (until 2056) were calculated. The specific objectives include:

-

determine the size of organic soil subsidence after 50 years of drainage (1966–2016) and the average annual rate of subsidence;

-

compare the field values for average annual subsidence rate (cm year⁻¹) across 50 years (1966–2016) with the subsidence rate values calculated using 14 empirical equations from the literature;

-

estimate the future rate of soil subsidence using 4 equations that include time since drainage as a factor;

-

use of the measured peat subsidence rate for estimation of groundwater level (using 4 empirical equations) and compare the obtained results with independent field measurements of groundwater level;

-

determine the average annual rate of organic matter mineralization and the percentage effect of chemical and physical processes on the total subsidence of drained organic soils.

In the study, a long period (1966–2016) of drainage was taken into account. Short periods may underestimate or overestimate the subsidence rate, as the mineralization processes at peatlands do not occur equally in time.

2. Materials and Methods

2.1. Study Area

The analyzed site—Solec Peatland (Łąki Soleckie)—has an area of approx. 220 ha and is located in central Poland (Figure 1) in the mesoregion designated as the Warsaw Plain in the geophysical division of Poland [51]. It is located near the village of Solec, Masovia Province, rural commune (district) of Góra Kalwaria (52°02′39″ N; 21°06′17″ E). The site is crossed by the River Mała, a left tributary of the River Jeziorka, which is itself a left tributary of the Vistula [19]. The average annual air temperature and precipitation are 8.2 °C and 519 mm, respectively (years 1971–2000). The average temperature during growing season for the period 1960–2017 was 13.96 °C, the average annual precipitation was 396.7 mm [52].

The organic soils in the area are made up of low sedge and reed-sedge peats, with medium to strong degree of decomposition [53]. Due to drainage, the top layer of the organic soil underwent a secondary transformation and is nowadays termed mursh layer [1,23]. Currently, the organic layer is the thickest (about 120–130 cm) at the central part of the site, along the River Mała. This soil is a hemic murshic according to the Polish Soil Classification (PSC) [23] and a Eutric Rheic Murshic Hemic Histosol according to the WRB 2015 [54]. Around the edge of the Solec peatland, organic soils gradually become more shallow (with the average thickness of the organic layer being approx. 30–40 cm). These are classified as thin sapric murshic soils according to the PSC but the WRB 2015 no longer classifies them as organic soils (Histosols) but rather Eutric Histic Gleysols (Drainic). Below the organic matter layer, there is a quaternary loose sand layer over 50 m thick [55].

The first drainage efforts at the site involved excavating a new bed for the Mała River, which commenced in 1941 and concluded in 1943. In 1967, a design of network of drainage and irrigation ditches spaced 60–130 m apart was developed [56]. The development of this project was preceded by preparatory works carried out by the Central Office of Water and Melioration Studies and Projects (Warsaw Branch) in 1966. The drainage project was developed according to the methodology commonly used in Poland in 1960–1990, when land reclamations of approx. 80% of fen areas (i.e., approx. 1 million ha) located in river valleys were carried out [1]. As part of the project, research was carried out at Solec peatland, including measurements of the peat deposit thickness and groundwater level at 19 points marked at the situational-altitude maps (scale 1:2000). Measurements of the peat deposit thickness were repeated several times in 1966 using a soil pin, while measurements of the groundwater level were carried out in observation wells using a hydrological whistle.

The reclamation works commenced in 1967 and divided the site into 13 plots (Figure 1). A two-way water regime (infiltration system) was used for the irrigation and drainage of the grasslands. The sub-irrigation system was equipped with 42 gated culverts and 37 regular culverts (bridges) used to move about the site [53]. For approx. 20 years, only a part of the site (northern part; Figure 1) has been used as intensively managed meadowland (approx. 40% of the area) and is mown 1–2 times annually without fertilization. The remainder is fallow land covered with nitrophilous tall-herbs, mostly stinging nettle (Urtica dioica), which gradually transition to areas overgrown with bushes and trees. In the eastern part of the Solec peatland, forest stands dominated by Alnus glutinosa and Betula sp. have formed through natural succession—a process that started at the time of World War II (Figure 2). There has been no water management at the site for approx. 20 years. The ditches are usually dry in the summer. The area has been a part of the Chojnów Landscape Park since 1993 and has been protected under the Natura 2000 network (PLH 140055) since 2011.

Given the above, the investigated area can be considered to be a typical central Europe peatland, i.e., fen peatland located in river valley, which was artificially drained and used as grassland at various intensity levels.

2.2. Field Work

Based on the situational-altitude maps (scale 1:2000) [56], 19 measurement points marked in 1966 were again located in 2016. The points are distributed evenly across the Solec peatland (Figure 1). At each measurement point, 3 benchmarks (stabilization points) were permanently installed to measure the thickness of the peat deposit and observation wells to gauge the groundwater level. Measurements of the peat deposit thickness were carried out using a soil pin, while measurements of the groundwater level were carried out in observation wells using a hydrological whistle. All measurements were made with an accuracy of 1 cm. The measurements were performed monthly from May to October 2016. The year in question (2016) showed slightly lower total precipitation during the growing season of Apr–Oct (334.7 mm) when compared against the data for 1960–2017 (396.7 mm). The months of May, June, July and September had comparatively lower precipitation totals than the corresponding months of the 1960–2017 period. By contrast, the precipitation total in October 2016 (115.7 mm) was almost three times higher than the average for 1960–2017 (39.8 mm) [51].

2.3. Data Processing

On the basis of archival materials contained in the land reclamation project on the thickness of the peat deposit and groundwater levels of the Solec site from 1966 and measurements of the thickness at the same measurement points during the 2016 growing season, verification of empirical formulas for the magnitude of peat soil subsidence in a 50-year perspective was undertaken. Based on the 1966 measurement points (land reclamation project), the peat deposit volume was re-measured after 50 years on several measurement dates during the 2016 growing season.

The field values and historical data were used to calculate: the size of Solec peatland surface lowering in 1966–2016, mean peat subsidence, percentage reduction in peat thickness and average annual rate of subsidence over this period. The data for peat deposit thickness in 1966, time since drainage (50 years) and depth of drainage ditches were fed into 14 empirical equations to calculate average annual rate of peat subsidence (Table 1).

Having a literature review of empirical equations for the rate of subsidence, a comparison was made between the results of calculating the magnitude of subsidence by empirical formulas and field measurements. The past measured field subsidence values (1966–2016) and empirical equations were used to calculate predicted average rates of peat subsidence in the forthcoming 20, 30 and 40 years (since 2016). Based on measured rates of peat subsidence, 4 empirical equations (Equations (15)–(18)) were used to estimate the depth to groundwater during growing period. The estimated groundwater levels in the vicinity of several measurement points were verified on the basis of independent historical daily measurements of groundwater level. These measurements come from several diploma theses, which are stored in the Warsaw University of Life Sciences Library. The above estimated groundwater levels were used to calculate the subsidence of peat surface due to oxidation (Equation (19)) and the volume of organic matter loss due to the mineralization process (Equation (20)). The application of the above equations made it possible to determine the percentage of chemical and physical processes responsible for the peat subsidence.

Additionally, the loss of peat per ha per year calculated by Jurczuk equation (Equation (20)) was compared to the measured organic matter (OM) loss per ha per year. In 1966 bulk density was between 0.168 g cm⁻³ and 0.240 g cm⁻³ (mean 0.186 g cm⁻³) and organic matter content was between 797 g kg⁻¹ and 878 g kg⁻¹ (mean 844 g kg⁻¹) [56]. Currently, the bulk density ranges between 0.25 g cm⁻³ and 0.40 g cm⁻³ and organic matter content between 659 g kg⁻¹ and 759 g kg⁻¹ (mean 709 g kg⁻¹) [57,58,59]. On the basis of the above data and peat thickness measured during field investigations, the stock of organic matter (OM stock) was calculated according to the formula:

OM stock [Mg ha⁻¹] = OM × BD × H,

where, OM is the mean organic matter content in the peatland [%], BD is the mean bulk density of peat deposit [g cm⁻³], H is the peat thickness [cm].

All calculations were performed using Statistica 13.3 software.

Table 1. Empirical equations for the second phase of peat surface subsidence by various authors.

Site (Source)	Equation (Number)		Explanations of Symbols Acc. to Sources
Depended on initial depth of deposit
Noteć River Valley; drainage intensity of peatlands: -low (0.4–0.6 m), -medium (0.6–1.0 m), -high (1.0–1.2 m), -total [60]	$y=0.051x+8.6$ $y=0.05x+18$ $y=0.082x+34.6$ $y=0.101x+9.5$	(1) (2) (3) (4)	y—surface subsidence [cm] x—initial depth of deposit [cm]
Noteć River Valley; drainage intensity of peatlands: -medium (0.6–1.0 m), -high (1.0–1.2 m), -total [60]	$y=0.00107x+0.34$ $y=0.00228x+0.47$ $y=0.0021x+0.17$	(5) (6) (7)	y—surface subsidence [cm year−1] x—initial depth of deposit [cm]
Peatlands in Central Europe [60]	$y=0.12x+23$	(8)	y—surface subsidence [cm] x—initial depth of deposit [cm]
Peatland of Moscow Reaserch Station [60,61]	$y=0.156x+19.2$	(9)	y—surface subsidence [cm] x—initial depth of deposit [cm]
Biebrza River Valley; Kuwasy fen [62]	$y=0.099x+2.9$	(10)	y—surface subsidence [cm] x—initial depth of deposit [cm]
Depended on years after drainage
Stepnica and Góra; fens [63]	$h=a·x^b$	(11) *	h—surface subsidence [cm] x—time since drainage [years] a, b—empirical coefficients
Various peatlands [45,64]	$S=5.9·{0.91}^y+1.2$	(12)	S—subsidence [cm year−1] y—time after drainage [years]
Depended on initial depth of deposit and years after drainage
Stary Borek; fen [13]	$S=1.88H^{0.204}·t^{0.6}$	(13)	S—subsidence [cm] H—initial depth of deposit [cm] t—time after drainage [years]
Depended on initial depth of deposit, years after drainage and the depth of the ditch
Noteć River valley; fen [60]	$h=0.14H+0.33t+0.005L-0.53$	(14)	h—surface subsidence [m] H—initial depth of peatland [m] t—depth of ditches [m] L—time after drainage [years]
Depended on the depth of ground water level
Zegvelderbroek; low-moor peat [65]	$Y=0.0281 X-0.581$	(15)	Y—predicted subsidence [cm year−1] X—average depth to water table [cm]
Poland; lowland fen peatland [13]	$S=0.024 D-0.54$	(16)	S—subsidence of soil surface [cm year−1] D—average depth of ground water in growing period [cm]
Zegveld; peatland [66]	$S=23.54 ALGL-6.68$	(17)	S—subsidence of soil surface [mm year−1] ALGL—average lowest groundwater level in summer [m]
High-latitude peatlands [5]	$Subs=-0.0212 WTD+0.43$	(18)	Subs—subsidence rate [cm year−1] WTD—mean water table depth [cm]
Equation for oxidative subsidence
Zegvelderbroek; low-moor peat [65]	$Y=0.0134 X-0.291$	(19)	Y—predicted subsidence [cm year−1] X—the average depth to water table [cm]
Poland; lowland fen peatland [13]	$M=-5.73+0.337D-0.925 DP$	(20)	M—loss of organic matter due to mineralization process [t ha−1 year−1] D—mean water table depth in growing period [cm] P—bulk density of peat deposit [g cm−3]

* 11A: a—9.90, b—0.346; 11B: a—6.64, b—0.576.

Table 1. Empirical equations for the second phase of peat surface subsidence by various authors.

Site (Source)	Equation (Number)		Explanations of Symbols Acc. to Sources
Depended on initial depth of deposit
Noteć River Valley; drainage intensity of peatlands: -low (0.4–0.6 m), -medium (0.6–1.0 m), -high (1.0–1.2 m), -total [60]	$y=0.051x+8.6$ $y=0.05x+18$ $y=0.082x+34.6$ $y=0.101x+9.5$	(1) (2) (3) (4)	y—surface subsidence [cm] x—initial depth of deposit [cm]
Noteć River Valley; drainage intensity of peatlands: -medium (0.6–1.0 m), -high (1.0–1.2 m), -total [60]	$y=0.00107x+0.34$ $y=0.00228x+0.47$ $y=0.0021x+0.17$	(5) (6) (7)	y—surface subsidence [cm year−1] x—initial depth of deposit [cm]
Peatlands in Central Europe [60]	$y=0.12x+23$	(8)	y—surface subsidence [cm] x—initial depth of deposit [cm]
Peatland of Moscow Reaserch Station [60,61]	$y=0.156x+19.2$	(9)	y—surface subsidence [cm] x—initial depth of deposit [cm]
Biebrza River Valley; Kuwasy fen [62]	$y=0.099x+2.9$	(10)	y—surface subsidence [cm] x—initial depth of deposit [cm]
Depended on years after drainage
Stepnica and Góra; fens [63]	$h=a·x^b$	(11) *	h—surface subsidence [cm] x—time since drainage [years] a, b—empirical coefficients
Various peatlands [45,64]	$S=5.9·{0.91}^y+1.2$	(12)	S—subsidence [cm year−1] y—time after drainage [years]
Depended on initial depth of deposit and years after drainage
Stary Borek; fen [13]	$S=1.88H^{0.204}·t^{0.6}$	(13)	S—subsidence [cm] H—initial depth of deposit [cm] t—time after drainage [years]
Depended on initial depth of deposit, years after drainage and the depth of the ditch
Noteć River valley; fen [60]	$h=0.14H+0.33t+0.005L-0.53$	(14)	h—surface subsidence [m] H—initial depth of peatland [m] t—depth of ditches [m] L—time after drainage [years]
Depended on the depth of ground water level
Zegvelderbroek; low-moor peat [65]	$Y=0.0281 X-0.581$	(15)	Y—predicted subsidence [cm year−1] X—average depth to water table [cm]
Poland; lowland fen peatland [13]	$S=0.024 D-0.54$	(16)	S—subsidence of soil surface [cm year−1] D—average depth of ground water in growing period [cm]
Zegveld; peatland [66]	$S=23.54 ALGL-6.68$	(17)	S—subsidence of soil surface [mm year−1] ALGL—average lowest groundwater level in summer [m]
High-latitude peatlands [5]	$Subs=-0.0212 WTD+0.43$	(18)	Subs—subsidence rate [cm year−1] WTD—mean water table depth [cm]
Equation for oxidative subsidence
Zegvelderbroek; low-moor peat [65]	$Y=0.0134 X-0.291$	(19)	Y—predicted subsidence [cm year−1] X—the average depth to water table [cm]
Poland; lowland fen peatland [13]	$M=-5.73+0.337D-0.925 DP$	(20)	M—loss of organic matter due to mineralization process [t ha−1 year−1] D—mean water table depth in growing period [cm] P—bulk density of peat deposit [g cm−3]

* 11A: a—9.90, b—0.346; 11B: a—6.64, b—0.576.

3. Results

3.1. Estimated Size and Rate of Peat Subsidence

Table 2. Mean ground water level, thickness of peat deposit (H) in 1966 and 2016, and mean annual rate of peat subsidence for individual study points on the Solec peatland.

Study Point No.	Mean Ground Water Level (cm)		Mean Thickness of Peat Deposit—H (cm)				Difference H(1966)—H(2016)	Reduction of H	Peat Subsidence Rate	Land Use
	1966	2016	1966	SD *	2016	SD	cm	%	cm Year−1
1	20 (±12)	45 (±13)	150	2.3	57	1.8	93	62.0	1.86	extensive grassland
2	20 (±10)	38 (±19)	60	1.8	56	2.2	4	6.7	0.08	herb vegetation **
3	35 (±13)	41 (±26)	100	2.4	56	1.1	44	44.0	0.88	herb vegetation
4	20 (±14)	61 (±18)	90	1.6	31	2.3	59	65.5	1.18	herb vegetation
5	35 (±17)	49 (±12)	150	2.9	83	1.6	67	44.7	1.34	herb vegetation
6	25 (±12)	52 (±15)	120	3.6	61	1.4	59	49.2	1.18	herb vegetation
7	70 (±15)	38 (±11)	120	1.8	105	2.5	15	12.5	0.30	herb vegetation
8	55 (±26)	41 (±14)	150	2.9	100	3.2	50	33.3	1.00	herb vegetation
9	80 (±19)	55 (±13)	70	2.4	49	1.9	21	30.0	0.42	extensive grassland
10	40 (±17)	29 (±17)	150	2.9	125	2.8	25	16.7	0.50	herb vegetation
11	120 (±24)	22 (±14)	180	1.8	171	2.0	9	5.0	0.18	herb vegetation
12	120 (±15)	55 (±16)	110	4.1	35	3.1	75	68.2	1.50	extensive grassland
13	60 (±24)	62 (±15)	100	3.8	13	1.8	87	87.0	1.74	herb vegetation
14	70 (±13)	45 (±16)	90	1.5	45	2.7	45	50.0	0.90	extensive grassland
15	90 (±24)	65 (±19)	150	3.5	40	3.2	110	73.3	2.20	herb vegetation
16	60 (±13)	71 (±24)	150	3.1	94	2.1	56	37.3	1.12	forest
17	60 (±22)	63 (±19)	150	4.1	49	1.7	101	67.3	2.02	forest
18	80 (±25)	46 (±21)	60	2.9	39	2.6	21	35.0	0.42	forest
19	70 (±18)	48 (±18)	70	1.8	44	2.9	26	37.1	0.52	forest
Mean			117		66		51	43.5	1.02

* SD—standard deviation; **—spontaneous herb vegetation on abandoned meadows developed since ca. 2000.

shows the average peat deposit thicknesses from 1966 to 2016, and groundwater levels at 19 measurement points. The field values for thickness indicate that the surface of the dewatered peat soil at different measurement points (Figure 1) subsided by a wide range of 4 cm to 110 cm (points 2 and 15) over 50 years, which was, respectively, 6.7% and 73.5% of the original thickness as measured in 1966. Land subsidence and peat thickness reduction averaged 51 cm and 43.5%, respectively. The average annual rates of subsidence over 50 years fell within the wide range of 0.08 cm year⁻¹ (point 2) to 2.2 cm year⁻¹ (point 15). Though both points are located near the Mała River (in 1967 transformed into a ditch) [54], their respective peat thickness values differed considerably in 1966 at 60 cm and 150 cm (points 2 and 15, respectively). The rate of subsidence and the peat thickness is mainly related to the groundwater level after drainage, as it is shown in Figure 3. In 1966 the thickness of peat and groundwater level did not show any relationships (Figure 3A), which changed after 50 years of permanent drainage, and in 2016 the linear relationship between the position of groundwater and peat thickness became clear (Figure 3B).

This means that lower groundwater level caused higher peat mineralization resulting in decreased peat thickness shown in Figure 3. Average annual rate of subsidence is also determined by the land use and plant cover for the given Solec peatland parts. The average annual rates of subsidence were: 0.96 cm year⁻¹ for fallow land overgrown with herb vegetation, 1.02 cm year⁻¹ for forested area, and 1.17 cm year⁻¹ for land used for extensive grassland. The Solec peatland was once used for agriculture, but parts of the land have since then been put to other land uses (partially due to the natural plant succession). As a result, the average annual rate of subsidence is fairly similar across these three types of land use (Table 2).

Table 3. Measured (19 study points) and calculated (with 14 empirical equations) mean annual rates of subsidence (cm year⁻¹) in the Solec peatland.

Study Point No.	Peat Loss (cm Year−1)	Ilnicki								Stankevič and Karelin	Krzywonos	Jurczuk		Maslov et al.	Jurczuk	Ilnicki
		1	2	3	4	5	6	7	8	9	10	11A	11B	12	13	14
1	1.86	0.32	0.51	0.93	0.49	0.50	0.81	0.48	0.82	0.85	0.35	0.77	1.25	1.25	1.09	0.42
2	0.08	0.23	0.42	0.79	0.31	0.40	0.61	0.29	0.60	0.57	0.17	0.77	1.25	1.25	0.91	0.17
3	0.88	0.27	0.46	0.85	0.39	0.45	0.70	0.38	0.70	0.69	0.25	0.77	1.25	1.25	1.00	0.28
4	1.18	0.26	0.45	0.83	0.37	0.44	0.67	0.35	0.68	0.66	0.24	0.77	1.25	1.25	0.98	0.25
5	1.34	0.32	0.51	0.93	0.49	0.50	0.81	0.48	0.82	0.85	0.35	0.77	1.25	1.25	1.09	0.42
6	1.18	0.29	0.48	0.88	0.43	0.47	0.74	0.42	0.75	0.76	0.29	0.77	1.25	1.25	1.04	0.34
7	0.30	0.29	0.48	0.88	0.43	0.47	0.74	0.42	0.75	0.76	0.29	0.77	1.25	1.25	1.04	0.34
8	1.00	0.32	0.51	0.94	0.49	0.50	0.81	0.48	0.82	0.85	0.35	0.77	1.25	1.25	1.09	0.42
9	0.42	0.24	0.43	0.81	0.33	0.41	0.63	0.32	0.63	0.60	0.19	0.77	1.25	1.25	0.93	0.20
10	0.50	0.32	0.51	0.94	0.49	0.50	0.81	0.48	0.82	0.85	0.35	0.77	1.25	1.25	1.09	0.42
11	0.18	0.35	0.54	0.98	0.55	0.53	0.88	0.55	0.89	0.95	0.41	0.77	1.25	1.25	1.13	0.50
12	1.50	0.28	0.47	0.87	0.41	0.45	0.72	0.40	0.72	0.73	0.27	0.77	1.25	1.25	1.02	0.31
13	1.74	0.27	0.46	0.85	0.39	0.45	0.70	0.38	0.70	0.70	0.25	0.77	1.25	1.25	1.00	0.28
14	0.90	0.26	0.45	0.84	0.37	0.44	0.67	0.36	0.67	0.66	0.24	0.77	1.25	1.25	0.98	0.25
15	2.20	0.32	0.51	0.94	0.49	0.50	0.81	0.48	0.82	0.85	0.35	0.77	1.25	1.25	1.09	0.42
16	1.12	0.32	0.51	0.94	0.49	0.50	0.81	0.48	0.82	0.85	0.35	0.77	1.25	1.25	1.09	0.42
17	2.02	0.32	0.51	0.94	0.49	0.50	0.81	0.48	0.82	0.85	0.35	0.77	1.25	1.25	1.09	0.42
18	0.42	0.23	0.42	0.79	0.31	0.40	0.61	0.30	0.60	0.57	0.18	0.77	1.25	1.25	0.91	0.17
19	0.52	0.24	0.43	0.81	0.33	0.41	0.63	0.32	0.63	0.60	0.20	0.77	1.25	1.25	0.93	0.20
Mean	1.02	0.29	0.47	0.88	0.42	0.46	0.74	0.42	0.74	0.75	0.29	0.77	1.25	1.25	1.04	0.33
	Conformity *	1	3	3	1	3	0	1	0	0	1	0	3	3	3	0

* number of calculated cases which are the closest (±10% of measured value) to the measured data, also marked in bold.

compares the average annual rate of subsidence as measured and as calculated with Equations (1)–(14) (Table 1). Three of the Equations (11A,B) and (12) produced identical figures for all measurement points, irrespective of the original thickness of the peat deposit. Comparing the results calculated by the equations with the measured values at study points, proved that the equations No. 2, 3, 5, 11B, 12 and 13 gave the results similar to the values measured in the field, i.e., not more than 10% deviation from the measured value. Even though Ilnicki’s equation (Equation (14)) considers three parameters (original/starting thickness of peat deposit, time since drainage, and depth of drainage ditches), the subsidence values produced by it differ greatly from those detected at the measurement points. Similarly, huge discrepancies were found for equations proposed by Stankevič and Karelin (Equation (9)) and Ilnicki (Equation (1)), the latter designed for shallow-drained deposits (0.4–0.6 m) as well as for equations proposed by Ilnicki (Equations (4) and (6)–(8)) and equations proposed by Krzywonos (Equation (10)) and by Jurczuk (Equation 11A). However, taking into account the values obtained for the whole site (19 study points), the results from the equation proposed by Jurczuk (Equation (13)) were almost identical (1.04 cm year⁻¹) to the average value from field measurements (1.02 cm year⁻¹). Based on the average values for the Solec peatland, annual rates of subsidence calculated by the equations differed and the conformity of the obtained results to field measurements can be ordered as follows: Equations (13), (2), (11B) and (12) then (11A), (9), (6) and (8). The results from other equations were far too distant from field measurements (Table 3).

3.2. Long-Term Subsidence Size Predictions

In addition to calculating the subsidence rates for the period 1966–2016 (Table 3), Equations (11A,B) and (12)–(14) were also used to estimate the future subsidence rate for the same soil (for the forthcoming 20, 30 and 40 years after 2016), assuming typical weather conditions for the area (

Table 4. Measured and calculated (1966–2016) mean rates of annual peat subsidence in the Solec peatland and predicted for the forthcoming 20, 30, and 40 years using empirical equations which take into account time after drainage from Table 1.

Period	Measured	Jurczuk (11A)	Jurczuk (11B)	Maslov et al., (12)	Jurczuk (13)	Ilnicki (14)
	(cm year−1)
1966–2016	1.02	0.69	1.25	1.25	1.03	0.33
1966–2036	-	0.61	1.08	1.21	0.91	0.23
1966–2046	-	0.56	1.02	1.20	0.86	0.21
1966–2056	-	0.52	0.97	1.20	0.82	0.18

Calculations based on the mean thickness of peat deposit in 1966—117 cm and in 2016—66 cm, as presented in Table 2.

). The subsidence rates measured for 1966–2016 (1.02 cm year⁻¹) were compared with those calculated for the same period. This comparison revealed that Jurczuk’s equation (Equation (13)) produces a value that closely matches the actual one (1.03 cm year⁻¹). The same equation applied to 2056 (40 year-period after 2016) shows that the rate of soil subsidence at the site will slow down by about 20%—from the current 1.03 cm year⁻¹ to approx. 0.82 cm year⁻¹. Notably, the subsidence rate estimate for 1966–2056, as calculated with Ilnicki’s equation (Equation (14)), was quite low at 0.18 cm year⁻¹, whereas the equation by Maslov et al. (Equation (12)) give larger value of 1.20 cm year⁻¹. All equations covered by this study (assuming stable climate and water conditions) predict that the average annual rate of subsidence will fall, but the extent of this change varies quite significantly across estimates—from approx. 3% to approx. 45% by 2056 (compared with the 1966–2016 measurements).

3.3. Estimates of Groundwater Level Based on Subsidence Rates

Equations (15)–(18) (Table 1), used for calculating subsidence rate, require input of the groundwater levels during the growing (summer) period. However, these equations had limited applicability, as no historical groundwater level data was available due to lack of consistent groundwater level monitoring or limited access to some plots (for example, some were overgrown with bush). Nevertheless, we did possess measurements of average annual subsidence rates, and attempted to estimate the groundwater table level on their basis (using the equations).

Table 5. Calculated mean ground water levels in the growing period based on measured annual peat subsidence rates using empirical equations No. 15–18.

Study Point No.	Peat Loss 1966–2016 (cm year−1)	Ground Water Levels (cm) during the Growing Period Calculated with Equations
		Schothorst (Equation (15))	Jurczuk (Equation (16))	Querner et al., (Equation (17))	Evans et al., (Equation (18))
1	1.86	86.8	100.0	107.4	108.0
2	0.08	23.5	25.8	31.8	24.0
3	0.88	52.0	59.2	65.7	61.8
4	1.18	62.7	71.7	78.5	75.9
5	1.34	68.4	78.3	85.3	83.5
6	1.18	62.7	71.7	78.5	75.9
7	0.30	31.3	35.0	41.1	34.4
8	1.00	56.3	64.2	70.8	67.4
9	0.42	35.6	40.0	46.2	40.1
10	0.50	38.5	43.3	49.6	43.8
11	0.18	27.1	30.0	36.0	28.8
12	1.50	74.1	85.0	92.1	91.0
13	1.74	82.6	95.0	102.3	102.3
14	0.90	52.7	60.0	66.6	62.7
15	2.20	98.9	114.2	121.8	124.0
16	1.12	60.5	69.2	75.9	73.1
17	2.02	92.5	106.7	114.2	115.6
18	0.42	35.6	40.0	46.2	40.1
19	0.52	39.2	44.2	50.4	44.8
Mean	1.02	56.9	64.9	71.6	68.3

shows the estimated groundwater levels during the summer season, calculated from Equations (15)–(18) on the basis of subsidence measurements at the 19 measurement points. Equations by Schothorst (Equation (15)) and Jurczuk (Equation (16)) gave similar values, ranging from approx. 57 to approx. 65 cm. Equations by Querner et al. (Equation (17)) and Evans et al. (Equation (18)) put the groundwater table at a slightly lower level (68–72 cm), which is the typical groundwater level during growing season in hay meadows on drained organic soils.

It also bears noting that the equations by Ilnicki (Equations (1)–(3), (5) and (6)) (Table 1) make distinctions of drainage intensity (for which drainage depths are given, which are closely related to the groundwater table level), and can also be used to indirectly estimate the groundwater level at the site. After comparing the field values with those calculated from Equations (1)–(3), (5) and (6) (Table 3), it was shown that Equations (3) and (5) produce results that quite closely follow the actual measurements, showing intensive (1.0–1.2 m) or medium-level (0.6–1.0 m) drainage at the site.

Groundwater levels were not consistently checked at the Solec site within the period concerned (1966–2016), but there have been some measurements at selected parts, spanning shorter periods.

Table 6. Comparison of calculated and measured ground water levels (1966–2016) using empirical equations No 15–18The text continues here (Figure 2 and Table 2).

Study Point No.	Calculated Ground Water Levels (cm) Using Equations				Measured Ground Water Levels (cm)	Reference
	Schothorst (Equation (15))	Jurczuk (Equation (16))	Querner et al., (Equation (17))	Evans et al., (Equation (18))
2	23	25	31	24	31 (Apr.-Sept. 2013–2015) *	[54]
3	52	59	65	61	71 (Jun.-Sept. 1976)	[66]
					40 (Jun.-Sept. 1977)
					56 (Jun.-Sept. 1978)
					54 (Jun.-Sept. 1979)
					60 (Jun.-Sept. 1980)
6	62	71	78	76	55 (Mar.-May 1970)	[67]
10	38	43	49	43	61 (Mar.-May 1970)	[67]
					57 (May-Aug. 1977)	[68]
					45 (Oct-Nov. 1981)	[69]
					42 (Apr.-Sept. 2013–2015)	[54]
11	27	30	36	28	24 (Apr-Sept. 2013–2015)	[54]
12	74	85	92	91	38 (May-Aug. 1977)	[68]

* in brackets month and year of measurement.

compares historical data for different months and years around 6 out of the 19 measurement points used in this study. This limited data indicates that the groundwater level estimates (Table 5) are a fairly close match for the direct measurements (Table 6). This means that using such equations (Equations (15)–(18)) to estimate the depth-to-groundwater (on the basis of the surface subsidence) may be useful way to recreate groundwater level data for a past period. However, more research is needed in this regard.

3.4. Estimating Mineralization of Organic Matter

The groundwater levels (Table 5) derived from one of Schothorst’s equations (Equation (15)) were then fed into another of Schothorst’s equations (Equation (19)), then assess the rate of subsidence caused by chemical processes (

Table 7. Assessment of the share of physical and chemical processes in mean annual peat subsidence rate using empirical equations of Schothorst (Equations (15) and (19)) and Jurczuk (Equations (16) and (20)).

Study Point No.	Peat Loss (cm year−1)	Ground Water Level (cm)	Subsidence (cm year−1)	Process (%)		Ground Water Level (cm)	Peat Loss (t ha−1 year−1)	Total Peat Loss (t ha−1 year−1)	Process (%)
		Schothorst (Equation (15))	Schothorst (Equation (19))	Chemical	Physical	Jurczuk (Equation (16))	Jurczuk (Equation (20))		Chemical	Physical
1	1.86	86.8	0.87	46.9	53.1	100.0	14.1	27.9	50.5	49.5
2	0.08	23.5	0.02	30.2	69.8	25.8	0.1	1.2	9.0	91.0
3	0.88	51.9	0.40	46.1	53.9	59.1	6.0	13.2	45.4	54.6
4	1.18	62.6	0.54	46.5	53.5	71.6	8.5	17.7	47.8	52.2
5	1.34	68.3	0.62	46.6	53.4	78.3	9.8	20.1	48.7	51.3
6	1.18	62.6	0.54	46.5	53.5	71.6	8.5	17.7	47.9	52.2
7	0.3	31.3	0.12	43.0	57.0	35.0	1.2	4.5	26.8	73.2
8	1.0	56.2	0.46	46.2	53.8	64.1	7.0	15	46.6	53.4
9	0.42	35.6	0.18	44.3	55.4	40.0	2.2	6.3	34.9	65.1
10	0.5	38.4	0.22	44.8	55.2	43.3	2.8	7.5	38.1	61.9
11	0.18	27.0	0.07	39.9	60.1	30.0	0.2	2.7	8.0	92.0
12	1.5	74.0	0.70	46.7	53.3	85.0	11.1	22.5	49.4	50.6
13	1.74	82.5	0.81	46.8	53.2	95.0	13.1	26.1	50.2	49.8
14	0.9	52.7	0.41	46.1	53.9	60.0	6.2	13.5	45.7	54.3
15	2.2	98.9	1.03	47.0	53.0	114.1	16.9	33	51.2	48.8
16	1.12	60.5	0.52	46.4	53.6	69.1	7.9	16.8	47.5	52.5
17	2.02	92.5	0.95	46.9	53.1	106.6	15.4	30.3	50.9	49.2
18	0.42	35.6	0.18	44.3	55.7	40.0	2.2	6.3	34.9	65.1
19	0.52	39.1	0.23	45.0	55.0	44.1	3.0	7.8	38.8	61.2
Mean	1.02	56.9	0.47	46.3	53.7	64.9	7.1	15.3	46.7	53.3

). With the field-measured average annual subsidence being 100% (the sum of physical and chemical processes), the size of subsidence determined from Equation (19) was expressed as the percentage effect of chemical processes. Chemical and physical processes were found to account for approx. 46% and 54% of the subsidence, respectively. Similar percentages were produced using the equations proposed by Jurczuk (Equations (16) and (20)). Equation (16) was used to estimate the groundwater table (Table 5), whereas Equation (20) served to calculate the annual loss of organic matter due to mineralization per ha. Soil mass loss per year per ha was calculated using the actual values for average annual rate of subsidence at different measurement points. Both types of soil loss were then compared to determine the extent to which chemical vs. physical soil processes contributed to subsidence. In both cases, using Equations (15) and (19) and Equations (16) and (20) produced similar results (Table 7).

The estimated loss of peat according to Jurczuk’s equation (Equation (20)) ranged between 0.1 t ha⁻¹ year⁻¹ and 16.9 t ha⁻¹ year⁻¹ (averaging 7.1 t ha⁻¹ year⁻¹). The 1966 study reported an average peat thickness of 117 cm (Table 2). Using the values of bulk density (BD) at 0.186 g cm⁻³, and organic matter content at 844 g kg⁻¹, the stock of organic matter was calculated to be 1836.7 t ha⁻¹. Given the current average peat thickness of 66 cm, BD of 0.325 g cm⁻³ and organic matter content of 709 g kg⁻¹, the present stock of organic matter amounts to 1520.8 t ha⁻¹ (Figure 4). This indicates that organic matter was lost at a rate of approximately 6.3 t ha⁻¹ year⁻¹, which is close to the value calculated with Equation (20), i.e., 7.1 t ha⁻¹ year⁻¹ (Table 7).

4. Discussion

Gathered data for peat thickness and subsidence related to the period 1966–2016. Groundwater levels from 1966 range between 20 and 120 cm, which indicates high variance in moisture levels at the site—from intensive dewatering to re-swamping. Where the depth-to-groundwater was 20 cm, further drainage would be required for agriculture use, which sparked the next phase of land subsidence. On the other hand, a depth-to-groundwater of 120 cm indicates intensive dewatering and mineralization of organic matter, again causing land subsidence (Table 2). This scenario is common for lowland fens with flat topography, where the variance in peat thickness stems from the relief of the mineral bottom [1,19,60]. The 2016 groundwater levels—ranging between 22 and 71 cm—indicate that the peat surface did indeed subside and draw closer attention to the groundwater level, which at some parts of the Solec peatland were also caused by lack of maintenance (deepening) of drainage ditches [19,70,71].

Oleszczuk et al. [19] estimated the annual rate of subsidence for a section of the Solec peatland (close to measurement point 2), and found that it averaged 0.62 cm year⁻¹ for the 20 years of meadow use (1978–1998) and the subsequent 20 years of fallowing. Notably, this value is lower than the one estimated for the peat as a whole in the present study (1.02 cm year⁻¹). The groundwater level and land use are inter-dependent and both factors influence the subsidence rate. The estimates of average annual rate of subsidence for drained peats may therefore differ, and in other countries used for forestry are as follows: [72]—0.37 cm year⁻¹, [73]—0.48 cm year⁻¹ (Finland), [5]—4.3 cm year⁻¹ (Indonesia). For meadow land, annual subsidence averages have been calculated by: [11]—0.50 cm year⁻¹ (Germany), [64]—0.53 cm year⁻¹ (The Netherlands), [74]—1.04–2.0 cm year⁻¹ (Norway), [17]—0.35–0.60 cm year⁻¹ (Poland), [39]—2 cm year⁻¹ (Ukraine), [75]—3 cm year⁻¹ (the Netherlands) and [76]—3.4 cm year⁻¹ (New Zealand).

The current state of research indicates that rates of organic matter mineralization in peat soils vary wildly across the world [25]. Findings for drained peatlands used for agriculture or forestry in boreal and temperate regions show that subsidence rates range from 0.002 to 0.12 m year⁻¹ [77]. The rate of subsidence is a function of multiple factors, including peat type, degree of peat decomposition, mineral matter content (ash content), bulk density, thickness of peat deposit, drainage depth, climate, land use, nutrient content and fertilization applied, duration of drainage [17,19,40,41,75].

Empirical annual rates of Solec peatland subsidence are most faithfully reproduced by equations of Jurczuk (Equation (13)) and Ilnicki (Equation (3)), designed for similar site conditions, i.e., temperate, central Europe lowland fens used as meadows [13,60]. There is a dearth of independent research focused on verifying empirical equations for calculating the size of surface subsidence in drained organic soils. Grzywna [17] tested 5 such equations—by Ostromęcki [10], Segeberg [78], Wertz [79], Mudd and Barret after [13] and Jurczuk [13]—on drained sites of eastern Poland across a period of 38 years (1974–2012), and found that field measurements were largely consistent with the values calculated with the equations. Equations that include temporal factors (Equations (11A,B) and (12)–(14)) can be used both to assess current peat subsidence, and to estimate future subsidence rates [80].

Peat subsidence and organic matter mineralization are mainly a function of drainage depth, type of peat, and drainage time. During the initial phase, which lasts several years (up to 15) after drainage, the peat body shrinks and compacts due to the falling moisture levels [42,81], whereas afterwards, mineralization of organic matter becomes the primary driver. Mineralization rates should be the highest immediately after dewatering and decrease through time [82]. In the initial phase of rapid subsidence (after drainage), the process is mostly driven by peat shrinkage, compaction, and consolidation. Later on, oxidation becomes the dominant factor, sometimes accounting for 90% of subsidence [42]. In the long-term (˃20 years of drainage) subsidence rates tend to stabilize and slow down [77,83]. Subsidence lowers the land surface and closes the distance to groundwater table. With the resultant higher moisture levels and slower oxidation of soil organic matter, subsidence slows down [84]. Due to next drainage episode, further lowering of the groundwater level took place, and the surface of peatland becomes more saturated with oxygen and decomposes further due to microbial processes, and consequently the peat surface lowers. These processes last as long as it takes for the soil surface to draw closer to the water surface, or alternatively, for the organic matter stock to become exhausted. It should also be noted that the changing moisture levels (for example, during the wetter seasons) can cause the peat to swell, i.e., slightly rise due to volume changes [35,85], but in the long-term studies such changes are negligible.

Peat consolidation, shrinkage, mineralization, swelling and subsidence are processes inherent to drained peatlands and linked to groundwater level. Even in the last century, mineralization was found to be primarily correlated with the groundwater level, so that mineralization can be ten times greater at lower groundwater levels [80,86]. Similar conclusions have been drawn from studies on greenhouse gas emissions. The emissions from grassland peatland, with mean annual water table < 20 cm below surface, was around 4 t CO₂–C–eq ha⁻¹ year⁻¹ and reached 7–9 t CO₂–C–eq ha⁻¹ year⁻¹ on soils with lower water table levels (of up to 39 cm below surface) [87]. It should also be noted that C loss caused by peat oxidation varies through time with changes in water table [82,88,89], a pattern corroborated by our study. Crucially, the theoretical values (calculated from equations) were similar to field measurements.

The oxidation constitutes from 30% to 85% of total subsidence in agriculturally used peatlands [33,65,74,76,90]. A study by Schothorst [65] on Netherlands peatland indicates that 20% of subsidence is attributable to irreversible shrinkage, 28% to consolidation and 52% to mineralisation. Another study, in New Zealand, estimated that compaction of drained peat soil accounts for 63% and soil organic matter (SOM) loss for 37% of subsidence [76]. Lipka et al. [38] found that 60% of total subsidence is due to oxidation and 40% due to compaction in Ukrainian peatlands. In the Scandinavian climate, in the case of peats drained for 50+ years, it is assumed that the two components of subsidence—compaction and SOM loss—have almost equal influence [74], which is also in line with our findings (chemical and physical processes were found to account for approx. 46% and 54% of the subsidence, respectively), indicating that physical and chemical processes in agriculturally used peatlands, irrespective of the regional factors, may have similar influence.

5. Conclusions

-

During 50 years the average lowering of peat deposits from 117 cm (1966) to 67 cm (2016), i.e., 43% was stated. The average annual rate of subsidence varied considerably across the site (from 0.08 to 2.2 cm year⁻¹). The average annual rates of subsidence were fairly similar across different types of land use: 0.96 cm year⁻¹ for fallow land overgrown with herb vegetation, 1.02 cm year⁻¹ for forest areas, and 1.17 cm year⁻¹ for grassland.

-

The applicability of the 14 equations was determined by comparing the average annual rate of subsidence measurements with the estimates (calculated with the equations). Values derived from Equations (3) and (11B) most closely matched actual field measurements. Therefore, these equations may be used in similar environmental conditions.

-

The 4 equations that included a temporal factor (time since drainage) were used to calculate both past and future subsidence rates. Equation (13) proved to be the most reliable, judging by the measurements and calculations for the period 1966–2016. When the current average annual rate of subsidence (1.03 cm year⁻¹) is fed into this equation, the resultant subsidence rate estimate for the year 2056 is 0.82 cm year⁻¹.

-

With the measured values of peat subsidence, it is possible to estimate the groundwater level during the growing season using empirical equations (Equations (15)–(18)), which was estimated at approx. 57–72 cm at Solec and these estimates were positively verified by historical field measurements.

-

Based on field measurements of peat subsidence at Solec (1966–2016) and the estimated groundwater level (Equations (15) and (16)), the share of chemical and physical processes in peat subsidence was determined. For this purpose, the empirical equation of Schothorst (Equation (19)) and Jurczuk (Equation (20)) were used. Both equations gave similar results, showing that the subsidence comprised approx. 46% of chemical processes and 54% of physical processes.

-

The loss of SOM at the Solec peatland, as a result of its mineralization, was estimated at 6–7 t year⁻¹ (Equation (20)), while the annual rate of loss of peat mass at approx. 15 t ha⁻¹ year⁻¹ as a result of chemical and physical processes resulting in peat subsidence.

-

The calculations and field results presented in the paper can be applied to central and western Europe. This is evidenced by similar results of calculating the amount of subsidence using equations developed especially for Dutch and Polish conditions.

-

Future research should be directed at further monitoring of the magnitude of peatland subsidence in terms of the climatic changes taking place (high air and soil temperatures, high evaporation values, low precipitation amounts). Such meteorological conditions are not conducive to the wet condition of these soils, especially shallow peat soils often located on mineral subsoil, which in periods of drought at low groundwater levels can additionally act as a drainage layer. In the long term, this can lead to the complete disappearance of peat soils from the environment.

-

Verification of the proposed equations for peatlands in other regions for which long-term studies are available would enrich the science with further research results.