• Ei tuloksia

1.5.1 Background

The study site is located in Maaninka (63˚09'49"N, 27˚14'3"E, 89 m above the mean sea level) in eastern Finland (Figure 1). Mean annual air temperature is 3.2°C in the region (reference period 1981-2010; Pirinen et al., 2012). Annual air temperature was about the long-term mean in 2009 (3.4°C), cooler in 2010 (2.0°C) and warmer in 2011 (4.4°C). The coldest month is February (-9.4°C) and the warmest July (17.0°C).

Mean annual precipitation is 612 mm (reference period 1981-2010; Pirinen et al., 2012). At the site, annual precipitation was less in 2009 (420 mm) and 2010 (520 mm) than the long-term mean but higher in 2011 (670 mm).

23 Figure 1 Location of the study site. (Map: Google, 2018).

The site is a 6.3 ha agricultural field. Prior to this experiment, it was cultivated with grass (Phleum pratense L.; Festuca pratensis Huds), barley (Hordeum vulgare L.) or oat (Avena sativa L.) during the last ten years. In addition, a pilot field campaign was carried out in 2008. The field was fertilized with dairy cow slurry (40 tons ha-1 containing 120 kg N ha-1, 19 kg P ha-1 and 112 kg K ha-1) in April 2008. Reed canary grass (RCG, Phalaris arundinacea L., variety ’Palaton’) was sown in mid-June 2008.

The seeds germinated poorly and plant growth was low. Therefore, glyphosate, a systemic herbicide, was applied in September and the field was ploughed in November 2008. The field was left fallow during winter 2008 – 2009.

1.5.2 Soil characteristics

The soil is classified as a Haplic Cambisol/Regosol (Hypereutric, Siltic) (IUSS Working Group WRB, 2007), the topsoil being generally silt loam based on the U. S.

Department of Agriculture (USDA) textural classification system. Other soil related variables are given in Table 1. Details of the analyses are given in Chapter 2.

22

1.4 LIFE CYCLE ASSESSMENT (LCA) OF BIOENERGY

Life cycle assessment (LCA) is a framework for estimation of environmental impact associated with a product within its life cycle. Based on the LCA ISO standard (ISO 14040:2006) the life cycle includes all steps from raw material production to disposal of the product in order to select the least burdensome options. However, the LCAs are adopted to focus on selected environmental impacts such as climate change, eutrophication or land use along the life cycle (Rebitzer et al., 2004). LCA is being increasingly used to estimate environmental impacts of agriculture. Based on the Web of Science search results for “life cycle assessment agriculture”, number of publications has increased from 38 in 1999 to 686 in 2017. The assessments have been carried out for farms (e.g. Haas et al., 2000; Beauchemin et al., 2010), bioenergy (e.g. Shurpali et al., 2010; Järveoja et al., 2013; Dias et al., 2017) and animal based products (e.g. Thomassen et al., 2008; Peters et al., 2010; Pelletier, 2018; Thévenot et al., 2018). However, comparison between LCAs is complicated due to lack of consistency in the analyses (Cherubini and Strømman, 2011; Caffrey and Veal, 2013).

The system boundaries for a life cycle assessment determine which processes and activities are considered in the overall LCA. Such analysis take into account the material and energy flows of primary processes, together with the extraction of raw materials. The use of a product, such as crop biomass of grassland, is also an important factor in the LCA. In general, LCA of cropping systems includes GHG balance of the cropping system, machinery and different energy inputs and outputs (Shurpali et al., 2010; Hakala et al., 2012; Järveoja et al., 2013).

1.5 SITE DESCRIPTION

1.5.1 Background

The study site is located in Maaninka (63˚09'49"N, 27˚14'3"E, 89 m above the mean sea level) in eastern Finland (Figure 1). Mean annual air temperature is 3.2°C in the region (reference period 1981-2010; Pirinen et al., 2012). Annual air temperature was about the long-term mean in 2009 (3.4°C), cooler in 2010 (2.0°C) and warmer in 2011 (4.4°C). The coldest month is February (-9.4°C) and the warmest July (17.0°C).

Mean annual precipitation is 612 mm (reference period 1981-2010; Pirinen et al., 2012). At the site, annual precipitation was less in 2009 (420 mm) and 2010 (520 mm) than the long-term mean but higher in 2011 (670 mm).

23 Figure 1 Location of the study site. (Map: Google, 2018).

The site is a 6.3 ha agricultural field. Prior to this experiment, it was cultivated with grass (Phleum pratense L.; Festuca pratensis Huds), barley (Hordeum vulgare L.) or oat (Avena sativa L.) during the last ten years. In addition, a pilot field campaign was carried out in 2008. The field was fertilized with dairy cow slurry (40 tons ha-1 containing 120 kg N ha-1, 19 kg P ha-1 and 112 kg K ha-1) in April 2008. Reed canary grass (RCG, Phalaris arundinacea L., variety ’Palaton’) was sown in mid-June 2008.

The seeds germinated poorly and plant growth was low. Therefore, glyphosate, a systemic herbicide, was applied in September and the field was ploughed in November 2008. The field was left fallow during winter 2008 – 2009.

1.5.2 Soil characteristics

The soil is classified as a Haplic Cambisol/Regosol (Hypereutric, Siltic) (IUSS Working Group WRB, 2007), the topsoil being generally silt loam based on the U. S.

Department of Agriculture (USDA) textural classification system. Other soil related variables are given in Table 1. Details of the analyses are given in Chapter 2.

24

Table 1 General soil characteristics (mean ± standard deviation) of top soil (0-18 cm).

Soil characteristic Mean ± SD

Clay (%) 25 ± 5.6

Silt (%) 53 ± 9.0

Sand (%) 22 ± 7.8

pH (H2O) 5.8 ± 0.19

Electrical conductivity (mS m-1) 14 ± 2.4 Soil organic matter (%) 5.2 ± 0.90 Particle density (g cm-3) 2.7 ± 0.014 Bulk density (g cm-3) 1.1 ± 0.11 Organic carbon (%) 3.0 ± 0.52 Total nitrogen (%) 0.2 ± 0.03

C/N 15 ± 0.40

K (mg dm-3 soil) 100 ± 13

P (mg dm-3 soil) 5.4 ± 1.3 Field capacity (%, soil moisture (v/v)) 40 ± 1.2 Wilting point (%, soil moisture (v/v)) 22 ± 0.80 1.5.3 Experimental design

The experimental design consisted of the main field and three experimental plots therein (Figure 2a). The main field was cultivated with reed canary grass (RCG, Phalaris arundinacea L., Figure 3a). Masts for eddy covariance (EC) measurements and for weather station were located in the middle of the field (Figure 2b). The net radiometer mast was located further away in order to avoid shading from the masts and instrument cabin. The blue hut in the middle of the field housed the EC gas analyzers.

Each of the three experimental plots were divided into three subplots (Figure 2c) and cultivated with either mixture of timothy and meadow fescue (TIM, Phleum pratense and Festuca pratensis, Figure 3c) or RCG. Third subplot was kept without any vegetation (BARE, Figure 3b). The order of the vegetated treatments was randomized. However, the BARE treatment was always in the middle of the vegetated subplots in order to prevent plant and root spread from one treatment to another. Permanent collars for gas flux measurements were installed in plot areas.

Sampling for soil and biomass were done within the plot areas.

25 Figure 2 (a) Aerial view of the experimental design (Picture: Perttu Virkajärvi, 2010) with closer view of (b) the main field set-up and (c) plots and subplot design at the study site.

Figure 3 (a) Reed canary grass, (b) soil without vegetation and (c) timothy and meadow fescue mixture at the study site.

24

Table 1 General soil characteristics (mean ± standard deviation) of top soil (0-18 cm).

Soil characteristic Mean ± SD

Clay (%) 25 ± 5.6

Silt (%) 53 ± 9.0

Sand (%) 22 ± 7.8

pH (H2O) 5.8 ± 0.19

Electrical conductivity (mS m-1) 14 ± 2.4 Soil organic matter (%) 5.2 ± 0.90 Particle density (g cm-3) 2.7 ± 0.014 Bulk density (g cm-3) 1.1 ± 0.11

Organic carbon (%) 3.0 ± 0.52

Total nitrogen (%) 0.2 ± 0.03

C/N 15 ± 0.40

K (mg dm-3 soil) 100 ± 13

P (mg dm-3 soil) 5.4 ± 1.3

Field capacity (%, soil moisture (v/v)) 40 ± 1.2 Wilting point (%, soil moisture (v/v)) 22 ± 0.80 1.5.3 Experimental design

The experimental design consisted of the main field and three experimental plots therein (Figure 2a). The main field was cultivated with reed canary grass (RCG, Phalaris arundinacea L., Figure 3a). Masts for eddy covariance (EC) measurements and for weather station were located in the middle of the field (Figure 2b). The net radiometer mast was located further away in order to avoid shading from the masts and instrument cabin. The blue hut in the middle of the field housed the EC gas analyzers.

Each of the three experimental plots were divided into three subplots (Figure 2c) and cultivated with either mixture of timothy and meadow fescue (TIM, Phleum pratense and Festuca pratensis, Figure 3c) or RCG. Third subplot was kept without any vegetation (BARE, Figure 3b). The order of the vegetated treatments was randomized. However, the BARE treatment was always in the middle of the vegetated subplots in order to prevent plant and root spread from one treatment to another. Permanent collars for gas flux measurements were installed in plot areas.

Sampling for soil and biomass were done within the plot areas.

25 Figure 2 (a) Aerial view of the experimental design (Picture: Perttu Virkajärvi, 2010) with closer view of (b) the main field set-up and (c) plots and subplot design at the study site.

Figure 3 (a) Reed canary grass, (b) soil without vegetation and (c) timothy and meadow fescue mixture at the study site.

26

1.5.4 Agricultural practices

The field was levelled before and after the sowing in June 2009 (Table 2). Barley was used as a protection crop on TIM and removed during the first harvest in August 2009. Germination gaps were filled on RCG areas in 2009. Mineral fertilizer was applied with the seeds in the first year and as a surface application during later years. Herbicide (Ariana-S, mixture of MPCA 200 g l-1, clopyralid 20 g l-1 and fluroxypyr 40 g l-1, 2 l with 200 l of water ha-1) was applied at the end of July 2009 to control the weeds on TIM and RCG. Biomass was harvested on TIM once in 2009 and twice during latter years on TIM (Table 3). RCG was harvested first time in spring 2011 to improve the crop growth and biomass quality for combustion (Burvall, 1997). On BARE, plants were removed weekly by hand.

Table 2 Sowing of the plants and corresponding seed rates (kg ha-1) on mixture of timothy and meadow fescue (TIM) and reed canary grass on main field (RCGmain) and at plot areas (RCGplot) in 2009.

Sowing

date Plant

(common name, Latin name, cultivar) Seed rate

TIM June 9 Timothy, Phleum pratense, Tuure 12

Meadow fescue, Festuca pratensis Huds., Antti 10

Barley, Hordeum vulgare, Voitto 120

RCGplot June 9 Reed canary grass, Phalaris arundinacea L., Palaton 11 RCGmain June 8 Reed canary grass, Phalaris arundinacea L., Palaton 11 Table 3 Use of fertilizers (kg ha-1), timing of harvesting events and corresponding yields (kg DW ha-1) on mixture of timothy and meadow fescue (TIM) and reed canary grass on main field (RCGmain) and at plot areas (RCGplot) during 2009 – 2011.

The main methods are described here shortly. Details of the methods, except for LCA, are given in the Chapters 2 through 5 and the references therein.

1.6.1 Climatic variables

Data logger (model: CR 3000, Campbell Scientific Inc.) was used to collect weather data (Table 4) at 30 minute intervals except for air pressure, which was collected hourly. Data was checked for quality and 30-min gaps were filled using linear interpolation. When data for air temperature, relative humidity, pressure or precipitation was missing for longer periods, data from nearby Maaninka weather station operated by Finnish meteorological institute (FMI) was used to fill the gaps.

Table 4 Weather station instruments. Measurement height of air temperature and net radiation sensor varied with the height of the EC mast due to plant height.

Variable Instrument Height (m)/Depth (cm)

Air temperature and

relative humidity HMP45C, Vaisala Inc 2.0, 2.4 or 2.5 m Photosynthetically

active radiation SKP215, Skye instruments Ltd. 2.0, 2.4 or 2.5 m Net radiation CNR1, Kipp&Zonen B.V. 2.0, 2.4 or 2.5 m Snow depth SR50A(h), Campbell Scientific Inc. 2.0, 2.4 or 2.5 m Wind speed and

wind direction 03002-5, R.M. Young Company 2.0, 2.4 or 2.5 m Rainfall 52203, R.M. Young Company 1 m

Air pressure CS106 Vaisala PTB110 Barometer 0.6 m Soil heat flux HPF01SC, Hukseflux 7.5 cm

Soil temperature 107, Campbell Scientific Inc. 2.5, 5, 10, 20 and 30 cm Soil moisture CS616, Campbell Scientific Inc. 2.5, 5, 10 and 30 cm 1.6.2 Greenhouse gas exchange

Closed-path eddy covariance method (EC, Baldocchi, 2003) was used to determine CO2 and N2O fluxes as well as the latent (LE) and sensible heat (H) fluxes of RCG cropping system at the main field. Instrumentation consisted of an infrared gas analyzer (IRGA) for CO2 and water vapor (H2O) concentrations (model: Li-7000 or Li-6262, LiCor), of a pulsed quantum cascade laser spectrometer for N2O, CO2 and H2O concentrations (Model: QC-Tildas-76-CS, Aerodyne Research Inc., USA) and a sonic anemometer (model: R3-50, Gill Instruments Ltd, UK) for wind velocity components and sonic temperature. The measurement heights were 2, 2.4 or 2.5 m depending on plant heights. Further details can be found from chapter 2 for CO2 and from chapter 3 for N2O and Rannik et al. (2015).

Season specific manual gas flux measurement methods were used in order to determine the annual GHG exchange of TIM, RCG and BARE in the plot areas. A snow-gradient method (Sommerfeld et al., 1993) was used during winter. Gas