Environmental parameters and management as factors affecting greenhouse gas emissions from clay soil

ABSTRACT

Greenhouse gas emissions (GHG) drive climate change, with agricultural land significantly contributing, influenced by soil properties. While extensive research exists on environmental and management impacts on GHG emissions across various soils and climates, understanding key factors influencing GHG emissions from clay soil in temperate climates is limited. This study aims to investigate the combination of environmental and management factors reducing N₂O, CO₂, and CH₄ emissions from clay soil in temperate climates. Recognising the potential of reduced tillage and legume-based crop rotations in mitigating GHG emissions, we investigate their impact on soil emissions. The conventionally managed field with spring barley, field beans, winter wheat, and winter rapeseed rotation demonstrates the lowest average N₂O emission (3.7 g N₂O ha⁻¹ d⁻¹), while the field with winter crops in a reduced tillage rotation shows the highest N₂O emission (8.5 g N₂O ha⁻¹ d⁻¹). A rotation with winter crops, beans twice, and barley, under conventional management, demonstrates the highest CO₂ emission (140.2 kg CO₂ ha⁻¹ d⁻¹), while the lowest CO₂ emission is observed in a rotation with winter crops, beans, and barley under reduced tillage management (100.8 kg CO₂ ha⁻¹ d⁻¹). CH₄ assimilation ranges from 3.1 to 5.4 CH₄ g ha⁻¹ d⁻¹ across all rotation and tillage combinations. However, ANCOVA results indicate that the volumes of GHG emissions are significantly influenced by the interaction of environmental and management factors, where precipitation is the most significant factor in the interaction with other environmental factors, soil tillage, and crop residues for N₂O and CO₂ emissions, while CH₄ emissions are influenced by the interaction of air temperature with other environmental factors, soil tillage, and crop residues. This underscores the need to consider both management and relevant environmental factors when evaluating the impact of practices on GHG emissions from clay soil in temperate climates.

KEYWORDS:

• Agriculture
• Cambic Calcisol
• field beans
• Picarro G2508
• spring barley
• tillage
• winter rapeseed
• winter wheat

Introduction

Human activities have caused global warming by increasing concentrations of nitrous oxide (N₂O), carbon dioxide (CO₂), and methane (CH₄) in the atmosphere (IPCC 2023). Land use change, unsustainable use of natural resources, energy consumption, as well as patterns of lifestyle and production, have contributed to the ongoing increase in global greenhouse gas (GHG) emissions over the last decade (IPCC AR6 SYR 2023). It is estimated that in 2019, the agri-food system, encompassing farm production activities, land use changes, and pre- and post- production processes, globally emitted 21% of CO₂, 53% of CH₄, and 78% of N₂O emissions (FAO 2021). More than 80% of all agricultural GHG emissions are attributed to CH₄ emissions from enteric fermentation and N₂O emissions from soils (EEA 2022). Therefore, introducing sustainable management practices is crucial for reducing the environmental impact of agriculture and ensuring the potential of productivity and incomes for farmers.

Soil can be both the source and the sink of GHG emissions. Microbial activity, root respiration, chemical decay processes, and heterotrophic respiration of soil fauna and fungi contribute to GHG emissions in soils (Chapuis-Lardy et al. 2007). The associated emission flux rates depend significantly on soil water content, soil temperature, nutrient availability, pH, and other land-use-related parameters (Oertel et al. 2016; Chataut et al. 2023). For instance, CH₄ production in soil occurs only under anaerobic conditions and is influenced by factors such as soil organic carbon content, soil temperature, and bulk density (Mitra et al. 2002). Environmental factors, including microbial populations, soil available carbon, soil nitrogen concentration, soil moisture, soil texture, soil temperature, soil pH, and the interaction of these factors with management, impact the rates of nitrification and denitrification, thereby affecting N₂O emissions from the soil (Wang et al. 2021b). The amount of CO₂ emitted from the soil is primarily determined by soil type, cropping system, climate, and soil organic carbon content (Bregaglio et al. 2022).

Soil management, encompassing soil preparation, pest, and weed control, represents crucial agro-technical measures to ensure the necessary conditions for plant growth. Conventional tillage, for example, aerates the soil surface to recover nutrients and reduce losses due to crop exports. However, long-term soil tillage can lead to soil compaction, increased decomposition of organic matter, a reduction in soil organic carbon, and elevated GHG emissions, affecting soil biota (Martin-Rueda et al. 2007; Peterson et al. 2019). Reduced tillage systems may mitigate these effects, offering several advantages compared to conventional tillage. With minimal mechanical impact, reduced tillage improves water infiltration, reduces soil erosion, enhances soil surface aggregates, increases soil organic matter and carbon content, moderates soil temperature, reduces production costs, and lowers CO2 emissions (Hobbs 2007; Peterson et al. 2019). However, this practice may increase the need for herbicides and pesticides as weed and pest control becomes more complex (FAO 2000). Overall, the impact of reduced tillage on GHG emission reduction is inconclusive, with some scientists claiming no difference in N₂O emissions between conventional and reduced tillage (Abdalla et al. 2013), while others conclude that reduced tillage increases N₂O emission (Badagliacca et al. 2018; Mei et al. 2018). While it is widely accepted that tillage practices lead to a decrease in soil organic carbon content over time (Lal 2003), other studies have also shown that soils under reduced tillage release more CO₂ compared to soils subjected to conventional tillage (Van den Bygaart and Angers 2006; Powlson et al. 2014; Shakoor et al. 2021). Tan et al. (2019) concluded that in dry years in Shandong Province, China, no-till practice along with straw incorporation and optimised fertilisation at a rate of 450 kg N ha⁻¹ yr⁻¹ may be a good strategy for achieving high yields while simultaneously reducing GHG emissions.

Current cropping systems in many regions worldwide are specifically designed for growing cereals with a significant reliance on synthetic nitrogen fertilisers (Plaza-Bonilla et al. 2017; Fiorini et al. 2020). Therefore, crop rotation may be considered a crucial long-term agricultural management strategy that enhances soil fertility, positively impacts yields, aids in pest and weed control, and increases the diversity of soil microorganisms (Venter et al. 2016). Crop rotation also has the potential to reduce GHG emissions from soil (Plaza-Bonilla et al. 2018; Liu et al. 2022). For instance, Rigon and Calonego (2020) suggest that in mesothermal climates, crop rotation improves the efficiency of soybean cultivation systems, potentially reducing CO₂ emissions through increased carbon and nitrogen input from crop residues, leading to higher soil organic carbon and carbon stocks. Meanwhile, Yang et al. (2023) found that, to increase soil organic carbon stocks and reduce GHG emissions without affecting yields, the most common crop rotation in a temperate semi-arid continental monsoon climate is a rotation with winter wheat, summer maize, and spring maize. In Finish study, crop rotation with legumes on clay soils emitted fewer GHG emissions compared to cereal crop monocultures (Lötjönen and Ollikainen 2017). Wheat grown after legumes exhibits higher yields with increased protein content in grains compared to wheat grown after other wheat. Legumes also increase the uptake of other nutrients by wheat, resulting in a generally healthier root system compared to monoculture (FAO 2016). When introducing legumes into rotation, the primary objective is to establish a connection with the requirements of the next crop. Organising crops in knowledge-based rotations promotes the efficient use of resources, improves soil quality, and may lead to a decrease in GHG emissions.

Thus, environmental parameters, along with information on land use and land management, are equally important factors in the formation of GHG emissions from soils. Despite numerous studies conducted in the last decade on the environmental factors and management practices affecting GHG emissions from different soil types in various climate zones, there is still no clear understanding of the main factors and combinations influencing N₂O, CO₂, and CH₄ emissions from clay soil in temperate climate. The combinations of environmental and management factors influencing GHG emissions from soil are also influenced by weather variability. Therefore, the aim of this study is to investigate which combination of environmental and management factors leads to decreased N₂O, CO₂, and CH₄ emissions from clay soil in a temperate climate. We hypothesise that reduced tillage and crop rotation with legumes decrease GHG emissions from clay soil in temperate climates.

Materials and methods

Experimental site

The experimental field is located at the Research and Study Farm (RSF) Peterlauki of the Latvia University of Life Sciences and Technologies (56°30.658′ N; 23°41.580′ E) (Figure 1). According to the IPCC guidelines, Latvia lies in a cool temperate moist climate zone (IPCC 2019), characterised by an average annual air temperature of +6.8°C and annual precipitation of 686 mm (LVĢMC n.d.). The soil type at the experimental field is a Cambic Calcisol with a silty clay texture, neutral pH, high organic matter content, and potassium levels, and medium in phosphorus levels (Valujeva et al. 2022). The average soil organic carbon content at a depth of 0-20 cm ranged from 1.68% to 2.44% in 2017 and 2022 (Supplementary Material Table S1). The agrochemical indicators of the soil are suitable for crop cultivation.

Figure 1. Location of the experimental field. Red numbers indicate the location of the experimental plots.

Trial design and treatments

The trial with two different tillage systems began in 2009 and comprises a total of 8 plots divided into 4 pairs, each plot measuring 25 × 100 m (0.25 ha) (Figure 1). The plots are located in flat terrain in close proximity to each other. The tillage systems applied in the experimental field are as follows: conventional tillage system (CT) involving mould-board ploughing to a depth of 24 cm; reduced tillage system (RT) with soil disc harrowing at a depth to 10 cm. Soil tillage occurred after harvesting the previous crop, and a soil compactor was used after CT to level the topsoil. In both tillage systems, one cultivation took place before sowing. The RT system involved soil disc harrowing performed twice after harvesting the previous crop.

From 2009 to 2013, the trial fields had four crop rotations, but since 2013, there has been a transition to two crop rotations: a three-year rotation with winter rapeseed (Brassica napus ssp. oleifera) followed by two years of winter wheat (Triticum aestivum) (WR–WW–WW) and a four-year rotation with the following crop sequence: field bean (Vicia faba)–winter wheat–rapeseed–spring barley (Hordeum vulgare) (FB–WW–WR–SB). In the spring of Year 5, winter rapeseed was sown with summer rapeseed due to meteorological conditions. Considering the carbon cycle in soil ranging from 20 to 60 years and acknowledging the importance of historical management (Trumbore 1997), the study distinguishes four crop rotations. This study presents results from the five growing seasons 2018–2022 (Year 1 to Year 5). An overview of plots in the experimental field is provided in Table 1.

Table 1. Crop rotations and tillage treatments for the eight experimental plots from Year 1 to Year 5.

Crop rotation	Tillage	Field	Year 1	Year 2	Year 3	Year 4	Year 5
1	RT	Plot 1	WW	WW	WR	WW	WW
	CT	Plot 2	WW	WW	WR	WW	WW
2	CT	Plot 3	WR	SB	FB	WW	SR
	RT	Plot 4	WR	SB	FB	WW	SR
3	RT	Plot 5	FB	WW	WR	SB	FB
	CT	Plot 6	FB	WW	WR	SB	FB
4	CT	Plot 7	SB	FB	WW	WR	SB
	RT	Plot 8	SB	FB	WW	WR	SB

The measurements were taken in plots marked in bold. Abbreviations: RT, reduced tillage; CT, conventional tillage; WW, winter wheat (Triticum aestivum); WR, winter rapeseed (Brassica napus ssp. oleifera); SB, spring barley (Hordeum vulgare); FB, field beans (Vicia faba); SR, spring rapeseed (Brassica napus).

Based on the agronomic needs of intensive crop production, a uniform set of agronomic measures were applied to each pair of plots. Nitrogen application rates in kilograms per hectare used for crop grown in plots from Year 1 to Year 5 are described in Table 2.

Table 2. Total rate of nitrogen (N kg ha⁻¹) applied for crop rotations from Year 1 to Year 5 including basic fertilisation for winter crops in autumn.

Crop rotation	Year 1	Year 2	Year 3	Year 4	Year 5
1	WW – N180	WW – N180	WR – N134	WW – N186	WW – N122
2	WR – N179	SB – N92	FB – N30	WW – N186	SR – N52
3	FB – N14	WW – N180	WR – N151	SB – N183	FB – N51
4	SB – N107	FB – N30	WW – N180	WR – N205	SB – N62

Abbreviations: WW, winter wheat (Triticum aestivum); WR, winter rapeseed (Brassica napus ssp. oleifera); SB, spring barley (Hordeum vulgare); FB, field beans (Vicia faba); SR, spring rapeseed (Brassica napus).

GHG emission measurements

Every two weeks from 10am to 2pm, soil flux measurements of N₂O, CO₂, and CH₄ were carried out from April to October of Year 1, Year 2, Year 3, Year 4, and Year 5. A Cavity Ring-Down Spectroscopy (CRDS) analyzer, Picarro G2508, was used in this study. The Picarro G2508 measures concentrations of N₂O, CH₄, and CO₂ emitted from the soil simultaneously with a one-second interval between measurements. The concentrations of N₂O, CO₂, and CH₄ from the soil were measured three times in each experimental plot at various places, with each measurement lasting 400 s, using non-transparent chambers with a diameter of 23 cm and a volume of 3 L (Valujeva et al. 2017). The chamber consists of a metal base with a sharpened lower edge, a non-transparent dome, and a sealing rubber between the base and the dome. The connection between chamber and Picarro G2508 is provided by a manufactured stainless-steel connector, a 9 m long Teflon tube with a diameter of 3.175 mm and an inner diameter of 1.587 mm, and a quick connector insulated with a rubber seal (Valujeva et al. 2017; Valujeva et al. 2022). The metal base was installed 30 min before the start of measurement, and the dome was laid on the base and connected to the analyzer just before the start of measurement. Also, the data logger Diver DI 500, Eijkelkamp, was placed in the chamber for the measurements on air temperature and pressure. The total number of measurements was 879. Furthermore, the Soil Flux Processor (SFP) software of Picarro Inc. was used for the calculations of emissions according to the linear model (Wagner et al. 1997). The data from SFP were converted to grams or kilograms per hectare per day.

The mean GHG budget in CO₂ equivalent was calculated for each crop and crop rotation by multiplying N₂O and CH₄ emissions by the corresponding global warming potential of AR4 (Forster et al. 2007) and adding CO₂ emissions. The mean GWP is calculated as tonnes per hectare per year.

Environmental variables

The data on air temperature, precipitation, humidity, and wind speed from the ‘Latvian Environment, Geology and Meteorology Centre’ meteorological observation station ‘Jelgava’ (56°33′24.954″ N and 23°57′50.679″E) were used to evaluate the weather conditions during the measurements. The average air temperature during the growing season of April 1 to October 31 was +14.9°C in Year 1, and +13.6°C, +13.3°C, +13.5°C, and +13.1°C in Year 2, Year 3, Year 4, and Year 5, respectively. The total precipitation during the growing season in Year 1, Year 2, Year 3, Year 4, and Year 5 was 336, 374, 435, 358, and 415 mm, respectively. The average humidity for the growing season in Year 1, Year 2, Year 3, Year 4, and Year 5 was 74.6%, 74.8%, 74.3%, and 75.9%, respectively. The average wind speed during the growing season in Year 1, Year 2, Year 3, Year 4, and Year 5 was 2.8, 3.3, 3.0, and 2.6 ms⁻¹, respectively.

Soil water content was measured during the GHG flux measurements using the Lutron PMS-714 Soil Moisture Meter, which measures soil water content at a depth of 10 cm. The average soil water content in the growing seasons of Year 1, Year 2, Year 3, Year 4, and Year 5 was 14.7%, 16.9%, 19.4%, 21.1%, and 20.4%, respectively. Additionally, soil temperature was measured during the GHG flux measurements using the Digital Temperature Meter Testo 922, which measures soil temperature in the upper layer of the soil. The average soil temperature in the growing seasons of Year 1, Year 2, Year 3, Year 4, and Year 5 was +20.6°C, +14.7°C, +16.1°C, +16.9°C, +20.7°C, respectively.

Statistical analysis

The data from the Soil Flux Processor (SFP) were combined into one dataset, along with weather data, soil water content, and soil temperature data, and subsequently analysed. Prior to analysis, the normality of GHG emissions from the soil was assessed using the Shapiro–Wilk test. For further analysis, non-parametric tests were applied (p < 0.05). A detailed description of the statistical analysis is provided in Table 3.

Table 3. A summary of the statistical analysis conducted in the study, including details on statistical methods employed, software’s used, and variables analysed.

Statistical analysis	Investigation	Software
Shapiro–Wilk test	Datasets distribution	IMB SPSS Statistics 22
Mann–Whitney U test	N2O emissions of tillage systems	IMB SPSS Statistics 22
Mann–Whitney U test	CO2, emissions of tillage systems	IMB SPSS Statistics 22
Mann–Whitney U test	CH4 emissions of tillage systems	IMB SPSS Statistics 22
Kruskal–Wallis test and Dunn’s post hoc tests	N2O emissions of tillage systems and crop rotations	IMB SPSS Statistics 22
Kruskal–Wallis test and Dunn’s post hoc tests	CO2 emissions of tillage systems and crop rotations	IMB SPSS Statistics 22
Kruskal–Wallis test and Dunn’s post hoc tests	CH4 emissions of tillage systems and crop rotations	IMB SPSS Statistics 22
Kruskal–Wallis test and Dunn’s post hoc tests	GWP of crops	IMB SPSS Statistics 22
Kruskal–Wallis test and Dunn’s post hoc tests	GWP of crop rotations	IMB SPSS Statistics 22
Kendall’s Tau correlation	Relationship between environmental variables and N2O, CO2, CH4 emissions	R Statistical Software version 3.2.5,
ANCOVA test	Effect of tillage systems, crop rotations and environmental variables on N2O, CO2, CH4 emissions	XL Stata

Results

Effect of tillage and crops on GHG emissions

Comparing tillage systems, CH₄ assimilation is statistically lower for reduced tillage (p = 0.043), while N₂O emission is significantly higher for reduced tillage (p = 0.027). However, there is no statistically significant difference in CO₂ emission between tillage systems (p = 0.754) (Table 4).

Table 4. Mean value and standard error of the mean (SE) for N₂O, CO₂ and CH₄ emissions of conventional tillage (CT) and reduced tillage (RT) for four crop rotations. Abbreviations: ¹, statistically significant difference with RT field with crop rotation 4; ², statistically significant difference with CT field with crop rotation 1; ³, statistically significant difference with CT field with crop rotation 3; ⁴, statistically significant difference with RT field with crop rotation 1; ⁵, statistically significant difference with CT field with crop rotation 2 (p<0.05).

Crop rotation	N2O (g ha^–1 day^-1)				CO2 (kg ha^–1 day ^-1)				CH4 (g ha^–1 day ^-1)
	CT		RT		CT		RT		CT		RT
	Mean	SE	Mean	SE	Mean	SE	Mean	SE	Mean	SE	Mean	SE
1	5.2	1.3	8.5	1.8	108.1	6.7	111.7	6.4	–5.4	0.6	–3.9^4	0.4
2	5.9	1.4	4.6	0.9	120.1	8.8	100.8	6.9	–4.25	0.5	–3.7^1,2	0.5
3	5.9	1.1	6.6	1.1	140.2	14.5	130.9	10.2	–4.6^3	0.7	–5.3^1	0.6
4	3.7	0.7	5.2	0.9	125.9	9.3	113.8	7.0	–3.7^1,2	0.5	–3.1^2,3,4,5	0.5
Total	5.2	0.6	6.1	0.6	122.1	4.8	112.5	3.7	–4.5	0.3	–3.9	0.3

By incorporating crop rotations into the analysis, the lowest average N₂O emission is observed in the conventionally managed plot with crop rotation 4, including field beans and spring barley since Year 1 (3.7 g N₂O ha⁻¹ d⁻¹) (Table 4). The crop rotation with winter crops and a reduced tillage system shows the highest N₂O emission (8.5 g N₂O ha⁻¹ d⁻¹). Crop rotation 3 under conventional management, featuring winter crops, beans twice in the rotation, and barley, demonstrates higher CO₂ emission compared to other rotations (140.2 kg CO₂ ha⁻¹ d⁻¹), while the lowest CO₂ emission is observed in crop rotation 2 with reduced tillage (100.8 kg CO₂ ha⁻¹ d⁻¹). All crop rotations show CH₄ assimilation ranging from 3.1 to 5.4 CH₄ g ha⁻¹ d⁻¹. Statistically significant differences are observed for CH₄ assimilation in both conventionally managed fields and fields with reduced tillage for different tillage system and crop rotation combinations. For instance, the CH₄ assimilation of the fields with reduced tillage and crop rotation 4 is significantly lower than other combinations, except with the conventionally managed field with crop rotation 4 and the field with reduced tillage and crop rotation 2 (p < 0.05) (Table 4).

The average GHG emission budget varies from 36.4 to 47.8 t CO₂ eq. per hectare per year across different crops and crop rotations, with no significant differences observed (Figure 2). Field beans have the highest average GHG emission budget, which is 18% higher than that of spring barley, 7% higher than winter wheat, and 10% higher than winter rapeseed. Conventional tillage with a crop rotation including winter crops shows a lower average GHG emission budget compared to other crop rotations. However, for reduced tillage, the lowest GHG emission budget is in crop rotation 2, where spring barley has been included twice in the crop rotation since Year 1.

Figure 2. GHG emission budget for (a) crops, (b) crop rotations calculated as global warming potential. Bars represent the means and the standard deviations of means. Abbreviations: CT, conventional tillage; RT, reduced tillage; WW, winter wheat (Triticum aestivum); SB, spring barley (Hordeum vulgare); FB, field beans (Vicia faba); WR, winter rapeseed (Brassica napus).

Almost all crops and crop rotations show a lower GHG emission budget for reduced tillage, except for spring barley with reduced tillage, which shows a higher global warming potential (GWP) compared to conventionally managed spring barley. Also, the plot with reduced tillage and a crop rotation with winter crops shows a higher GWP compared to conventionally managed crop rotation with winter crops. Additionally, when we calculated the GHG budged without CO₂ emission, we found that plots with reduced tillage produced significantly up to 21% more emissions compared to plots with a conventional tillage system (p = 0.015).

Relationship among the environmental variables and GHG emissions

A Kendall’s Tau correlation coefficient was computed to assess the relationship between N₂O, CO₂, CH₄ emissions, and environmental variables (p < 0.05). The highest correlation coefficient is observed between CO₂ and CH₄ emissions, ranging from −0.247 for reduced tillage to −0.335 for conventional tillage (Figure 3). N₂O emission shows a positive weak correlation with precipitation and wind speed, while it exhibits a negative weak correlation with CH₄ emission, air temperature, and soil temperature, with correlation coefficients ranging between −0.107 and 0.022. CO₂ emission demonstrates a moderately strong negative correlation with CH₄ emission and a weak negative correlation with wind speed and soil moisture. However, it has a weak positive correlation with relative humidity, with correlation coefficients ranging between −0.335 and 0.036. CH₄ emission shows a weak negative correlation with air temperature and soil temperature, but conventionally managed fields exhibit a weak positive correlation between CH₄ emission and precipitation. Environmental variables affect N₂O, CO₂, and CH₄ emissions from the soil differently, and there is no single environmental variable that affects all emissions equally.

Figure 3. Correlation plot among the environmental variables and GHG emissions: (a) for whole dataset, (b) for conventional tillage (CT); (c) for reduced tillage (RT). Numbers indicate the significant correlation coefficients, while the intensity of the colour of the boxes represents the level of correlation according to the scale (p < 0.05). Abbreviations: N2O, N₂O emission; CO2, CO₂ emission, CH4, CH₄ emission; Prec, the sum of the 5-day precipitation before the measurement (mm); AirT, 5-day average air temperature before the measurement (^oC); Hum, 5-day average relative humidity (%); Wind, 5-day average wind speed before the measurement (m/s); STemp, soil temperature during the measurement (^oC); SMoist, soil water content during the measurement (%).

GHG emissions are influenced by the interaction of several factors, encompassing both agro-technical and environmental aspects. To analyze the effects of agro-technical measures on GHG emissions from soil, ANCOVA was used, incorporating six quantitative variables: precipitation, air temperature, relative humidity, wind speed, soil temperature, and soil water content. The qualitative variable consisted of eight groups of agro-technical measures based on two soil tillage groups and four crop rotation groups. Three-level interactions were examined using the ANCOVA model.

N₂O emissions from clay soils are significantly influenced by relative humidity (p = 0.001), the interaction of precipitation, wind, and management factor (p = 0.000), and the interaction of wind, soil moisture, and management factor (p = 0.010). Using the Best model variables selection method, three variables have been retained in the model. Based on the Type III sum of squares, the following variables significantly explain the variability of the dependent variable N₂O: Hum; Prec * Wind * TILL_CR; Wind * SMoist * TILL_CR (Table 5). Among the explanatory variables, based on the Type III sum of squares, precipitation is the most influential factor for N₂O emission from soil.

Table 5. Summary of the most significant factor interactions influencing N₂O emissions from soil using ANCOVA (p < 0.05).

Factor	Dependent variable N2O emission
	Sum of squares	df	Mean square	F	p-value
Hum	1634.573	1	1634.573	10.277	0.001
Prec*Wind*TILL_CR	4567.015	8	570.877	3.589	0.000
Wind*SMoist*TILL_CR	2949.093	7	421.299	2.649	0.010

Abbreviations: Hum, 5-day average relative humidity (%); Prec, the sum of the 5-day precipitation before the measurement (mm); Wind, 5-day average wind speed before the measurement (m/s); TILL_CR, factor which represents tillage and crop rotation.

Relative humidity shows a statistically significant positive effects on N₂O emission from the soil (p = 0.001). The agro-technical measures in interaction with wind speed and soil moisture (crop rotation 4 with reduced tillage fixed at 0) have a positive effect on N₂O emissions from crop rotation 1, 3, and 4 with conventional tillage, and crop rotation 1 and 3 with reduced tillage. Meanwhile, agro-technical measures in interaction with precipitation and wind speed (crop rotation 4 with reduced tillage fixed) increase N₂O emission from crop rotation 2 with conventional tillage and crop rotations 1 and 2 with reduced tillage. Other agro-technical groups show a negative effect on N₂O emission from the soil. Results of the ANCOVA model for N₂O emission are found in Supplementary Material Table S2.

The CO₂ emission from clay soils is significantly influenced by the interaction of relative humidity and soil temperature, the interaction of precipitation, wind speed, and agro-technical measures, and the interaction of precipitation, soil temperature, and agro-technical measures (p < 0.05). Using the Best model variable selection method, three variables have been retained in the model. Based on the Type III sum of squares, the following variables significantly explain the variability of the dependent variable CO₂: Hum*Stemp; Prec*Wind*TILL_CR; Prec*STemp*TILL_CR (Table 6). Among the explanatory variables, based on the Type III sum of squares, precipitation is the most influential.

Table 6. Summary of the most significant factor interactions influencing CO₂ emissions from soil using ANCOVA (p < 0.05).

Factor	Dependent variable CO2 emission
	Sum of squares	df	Mean square	F	p-value
Hum*STemp	316954.637	1	316954.637	45.336	<0.0001
Prec*Wind*TILL_CR	642804.631	8	80350.579	11.493	<0.0001
Prec*STemp*TILL_CR	318344.558	8	39793.070	5.692	<0.0001

Abbreviations: Hum, 5-day average relative humidity (%); STemp, soil temperature during the measurement (oC); Prec, the sum of the 5-day precipitation before the measurement (mm); Wind, 5-day average wind speed before the measurement (m/s); TILL_CR, factor which represents tillage and crop rotation.

Relative humidity in interaction with soil temperature shows statistically significant positive effects on CO₂ emission (p < 0.0001). Agro-technical measures in interaction with precipitation and wind speed show a negative effect on CO₂ emission from crop rotation 2 with conventional tillage and reduced tillage (crop rotation 4 with reduced tillage is fixed). But agro-technical measures in interaction with precipitation and soil temperature show a positive effect on CO₂ emission from crop rotation 2 with conventional tillage and reduced tillage. Other agro-technical groups show the opposite or neutral effect on CO₂ emission from soil. Results of the ANCOVA model for CO₂ emission are found in Supplementary Material Table S3.

The CH₄ emission from clay soils is significantly influenced by the interaction of air temperature and agro-technical measures, the interaction of air temperature, soil moisture, and agro-technical measures, and the interaction of wind speed, soil temperature, and agro-technical measures (p < 0.0001) (Table 7). Using the Best model variables selection method, three variables have been retained in the model. Based on the Type III sum of squares, air temperature is the most influential. Results of ANCOVA model for CH₄ emission are found in Supplementary Material Table S4.

Table 7. Summary of the most significant factor interactions influencing CH₄ emissions from soil using ANCOVA (p < 0.05).

Factor	Dependent variable CH4 emission
	Sum of squares	df	Mean square	F	p-value
AirT*TILL_CR	3442.308	8	430.289	15.575	<0.0001
AirT*SMoist*TILL_CR	3207.059	7	458.151	16.584	<0.0001
Wind*STemp*TILL_CR	1421.409	8	177.676	6.431	<0.0001

Abbreviations: AirT, 5-day average air temperature before the measurement (^oC); SMoist, soil water content during the measurement (%); STemp, soil temperature during the measurement (^oC); Wind, 5-day average wind speed before the measurement (m/s); TILL_CR, factor which represents tillage and crop rotation.

Air temperature, in interaction with agro-technical measures, shows a positive effect on CH₄ emission from crop rotation 1 with conventional tillage, but other agro-technical measures show the opposite effect (crop rotation 4 is fixed). Air temperature, in interaction with soil moisture and agro-technical measures, shows a positive effect on CH₄ emission from all agro-technical measures (crop rotation 4 is fixed). While wind speed, in interaction with soil temperature and agro-technical measures, shows a positive effect on CH₄ emission from crop rotation 2 and 3 with conventional tillage (crop rotation 4 is fixed).

Discussion

Factors affecting GHG emissions from clay soil

Reducing soil tillage has been suggested to potentially decrease N₂O emissions from the soil (Pu et al. 2022); however, an opposite effect have also been observed, where reduced soil tillage promotes N₂O emissions (Yue et al. 2023). Our findings in Table 4 also indicate that, across all crop rotations, reduced tillage generally results in higher N₂O emissions, except for crop rotation 2. Nitrogen availability is recognised one of the most important factors influencing N₂O emissions from soil (Butterbach-Bahl et al. 2013), with soil moisture playing a critical role in regulating the production and release of N₂O emissions from soil (Imer et al. 2013). However, Figure 3 indicates no significant relationship between N₂O emission and soil moisture, suggesting that other factors may have a more dominant influence on the N₂O formation process from clay soils in a temperate climate. Additionally, our analysis identifies wind as a factor influencing N₂O emission from the soil (Table 5). While wind itself may not directly impact emission formation processes, it can indirectly affect N₂O emissions by accelerating soil drying through increased evaporation, thereby reducing soil microbial activity and leading to decreased N₂O emission (Smith et al. 2008). Furthermore, wind may enhance soil aeration and oxygen supply, influencing the balance of N₂O production and consumption (Venterea et al. 2012).

In Table 2, the data illustrates that increased CO₂ emissions associated with reduced tillage are specifically noticeable in crop rotations involving winter crops. Conversely, in other crop rotations, the pattern is reversed, although the difference is not statistically significant. Additionally, Yue et al. (2023) has affirmed that reduced tillage has no impact on CO₂ emissions, whereas adopting no-tillage practices leads to a significant 15.1% reduction in CO₂ emissions. CO₂ emission from soil is affected by a complex interplay of various biotic, abiotic, and anthropogenic factors, and as the most important of them are mentioned soil type, cropping system, weather conditions, and soil organic content (Bregaglio et al. 2022). In our study, relative humidity, soil temperature, precipitation, wind and management practices are the most important interacting factors affecting CO₂ emissions (Table 6).

Factors considered in the study interact in complex ways, and the magnitude of the influence on CH₄ emission from soil can vary depending on local environmental conditions and land management practices. Our study show that well-aerated soil has higher CH₄ assimilation rate, which is in line with Wang et al. (2021a), while there is also opinion that mechanical tillage stimulates the CH₄ emission from soil (Peterson et al. 2019). We found that an increase in air temperature and soil temperature reduces CH₄ emissions for both conventional and reduced tillage systems (Figure 3), which is contrary to the findings of Jiang et al. (2010), in alpine meadows. Both findings confirm the significant influence of local conditions and land management on GHG emissions from the soil.

The importance and planning of crop rotation and tillage system to reduce GHG emissions

Crop rotation is considered a sustainable farming practice that may help improve soil health that improve pest control, and potentially contribute to GHG emissions reduction through improved soil management. The Common Agricultural Policy (CAP) 2023–2027 determines that crop rotation or diversification has to be applied to 86% of the EU arable land (EC 2022). Crop rotation is included in 24 CAP Strategic plans, and all countries have allowed that the change of the main crop must be ensured every three years. But, for instance, Estonia has included an exception stating that a farm with an area less than 10 ha is not required to implement crop rotation (EC 2023). Renwick et al. (2021) found that crop rotation diversification also reduced stress on crops and yield reduction caused by climate change. Diversifying the maize-soybean rotation with cereals and cover crops increased drought resistance of maize and reduced yield losses by 17.1%. It is recommended not to grow legumes in the same field for several years in a row to minimise pest, disease and weed problems (Döring 2015). The experimental field with crop rotations 2, 3, and 4 has a four-year rotation cycle and field beans are followed by winter wheat, which is also considered good agricultural practice (Döring 2015). GHG emissions vary between both crop rotations and tillage systems, meaning that management factors need to be considered alongside the most relevant environmental factors for each GHG to determine the impact of a particular management practice on GHG emissions from clay soil.

Because of the ability of legumes to fix aerial nitrogen, supply high-quality protein, reduce the life cycle of soil-borne diseases, and provide nectar and pollen resources, legumes are considered as one of the beneficial crops in crop rotation (Döring 2015). Therefore, if the primary goal of including legumes in crop rotation is to improve soil health, little to no additional nitrogen fertiliser is needed. However, if the goal is high yield, a moderate amount of nitrogen fertiliser may be considered. We observed that in the year when beans are sown, there is a statistically significant decrease in N₂O emissions compared to the year before beans in the crop rotation. Also, in the year after beans, N₂O emission is statistically significantly lower than in the year when beans are sown. This could be explained by both the reduced amount of used nitrogen mineral fertilisers and the symbiosis with soil bacteria.

However, another study observed higher GHG emissions in years when soybeans were included in the rotation, explained by lower nitrogen utilisation efficiency compared to grain crops (Virk et al. 2022). Additionally, we found that the year before field beans and in the year when field beans are sown, N₂O emissions are lower for reduced tillage system. Still, in the year after field beans in crop rotation, reduced tillage shows higher N₂O emission compared to conventional tillage. CO₂ emission is lower in the year before field beans, but higher CH₄ assimilation is observed in the year when field beans are sown. This suggests that the GHG measurements so far allow us to understand that the inclusion of field beans in crop rotation is an advisable measure for reducing N₂O emission from clay soils. However, additional research is needed to investigate the effects of field beans on reducing CO₂ emissions, increasing carbon sequestration and CH₄ assimilation.

Assessment of management practices for climate change mitigation

Various factors, such as land use, vegetation, nutrient availability, soil moisture, precipitation, soil temperature, affect GHG emissions from soil (Oertel et al. 2016; Chataut et al. 2023). Our previous studies in a temperate climate have shown that the amount of GHG emissions from clay soils is influenced by differences in meteorological conditions between years (Valujeva et al. 2022). Additionally, Behnke and Villamil (2019) have investigated that year has a significant main effect on the formation of all GHG emissions from soil. Therefore, to assess the impact of a management practice on reducing GHG emissions, long-term observations of GHG emissions are needed, covering as wide a range of weather conditions as possible (Smith et al. 2008; Venterea et al. 2012; Smith et al. 2018). Reducing GHG emissions is a complex process because many factors influence the formation of GHG emissions, which vary over time and space. One solution may not be suitable for all soil types in all climate zones, as even within one field, the soil is heterogeneous.

Future research should focus on the development of innovative sensors and remote sensing techniques that can provide high-resolution data at large spatial scales. This includes the use of satellite instruments, drones, and ground sensors with improved accuracy and sensitivity. Long-term datasets are critical for tracking changes in soil GHG emissions over time. Future research should prioritise establishing and maintaining monitoring networks that span decades, allowing scientists to identify trends and differences in emissions under different land use change and climate scenarios. A deeper understanding of soil microbiological communities and their role in formation of GHG emissions is necessary. The impact of microbial diversity and activity on GHG production and consumption should also be investigated. This knowledge may inform strategies for manipulating microbial communities to reduce GHG emissions from soil. The development of robust models and forecasting tools is essential for predicting GHG emissions from soil under different scenarios. Future research should focus on improving existing models and incorporating new data sources, such as genetic and metagenomic information, to improve prediction accuracy. Further research into the effectiveness of soil management practices such as tillage, reduced tillage, and precision nutrient management, applied on different soil types (e.g. sandy and loamy soils), in reducing GHG emissions is essential. This research may help policy makers and farmers adopt sustainable agricultural practices. In research on GHG emissions from soil, attention should be paid not only to scientific aspects but also to policy development and to knowledge transfer between science, policy and practice. Close collaboration between scientists, policy makers, and the practitioners is necessary to ensure that research findings lead to meaningful action and behavioural change.

Conclusions

The combination of reduced tillage and crop rotation with spring barley and field beans show lower global warming potential, while the reduced tillage system with crop rotation involving only winter crops shows a higher global warming potential. However, apart from management factors, it is crucial to consider the most relevant environmental factors to determine the impact of management practices on N₂O, CO₂, and CH₄ emissions from clay soil in a temperate climate. Precipitation stands out as the most influential factor in the interaction with management, wind, and soil moisture, affecting N₂O emission from soil. Similarly, for CO₂ emissions, precipitation in interaction with soil temperature, wind, and management is the most significant factor. The dominant factor influencing CH₄ emissions is air temperature, especially when considered in interaction with soil moisture and management. Considering the previously investigated concept that annually variations have a significant impact on GHG emissions from clay soil, and now identifying the interaction of management factors and the most significant environmental factors for each gas, it is necessary to continue measurements. This is crucial to ensure that the identified factors are not influenced by annual variations and to include other soil types and management practices, such as sandy and loamy soils and organically managed areas.

Acknowledgments

The authors would like to thank to personnel of Study farm Peterlauki of Latvia University of Life Sciences and Technologies on cooperation performing measurements on a broad multifaceted field experiment.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Correction Statement

This article has been corrected with minor changes. These changes do not impact the academic content of the article.

Additional information

Funding

The research was funded by ‘Improvement of accounting system and methodologies for estimation of greenhouse gas (GHG) emissions and CO₂ removals from croplands and grasslands’ of the Ministry of Agriculture of the Republic of Latvia No.10 9.1-11/18/865-E, No.10 9.1-11/19/1748-E, No.10 9.1-11/20/1657-E, No. 10 9.1-11/21/1825-E, and No.0.9.1-11/22/1107-e.

Notes on contributors

Kristine Valujeva

Kristine Valujeva is a Leading Researcher with Scientific Laboratory of Forest and Water Resources at Latvia University of Life Sciences and Technologies. Her research interests include soil, sustainable land management, environmental governance and agricultural policies.

Jovita Pilecka-Ulcugaceva

Jovita Pilecka-Ulcugaceva is a Scientific Assistant with Scientific Laboratory of Forest and Water Resources at Latvia University of Life Sciences and Technologies and concurrently a doctoral candidate at Latvia University of Life Sciences and Technologies. Her research interests include air pollution, soil, and GHG emissions in agriculture.

Madara Darguza

Madara Darguza is a Researcher with Institute of Soil and Plant Science at Latvia University of Life Sciences and Technologies and concurrently a doctoral candidate at Latvia University of Life Sciences and Technologies. Her research interests include soil tillage, crop rotation, crop production.

Kristaps Siltumens

Kristaps Siltumens is a Scientific Assistant with Scientific Laboratory of Forest and Water Resources at Latvia University of Life Sciences and Technologies and concurrently a master student at Latvia University. His research interests include soil and waste management.

Ainis Lagzdins

Ainis Lagzdins is a Leading Researcher with Institute of Landscape Architecture and Environmental Engineering and a Head of Scientific Laboratory of Forest and Water Resources at Latvia University of Life Sciences and Technologies. His research interests include nutrient management, water quality, hydrology, drainage, water management.

Inga Grinfelde

Inga Grinfelde is a Scientific Assistant with Scientific Laboratory of Forest and Water Resources at Latvia University of Life Sciences and Technologies. Her research interests include soil, mitigation of GHG emissions in agriculture, hydrology, and water management.