Global Land Use Impacts on Biodiversity and Ecosystem Services in LCA

electronic supplementary material

Global Land Use Impacts on Biodiversity and Ecosystem Services in LCA

Land use impacts on freshwater regulation, erosion regulation and water purification: A spatial approach for a global scale level

Rosie Saad • Thomas Koellner • Manuele Margni

Received: 24 May 2012 / Accepted: 19 March 2013

© Springer-Verlag 2013

Responsible editor: Roland Geyer

R. Saad () • M. Margni

CIRAIG, Chemical Engineering Department, École Polytechnique de Montréal, P.O. Box 6079, Montréal, Quebec, H3C 3A7, Canada

e-mail:

T. Koellner

Faculty of Biology, Chemistry and Geosciences, Professorship of Ecological Services PES, University of Bayreuth, 95440 Bayreuth, Germany

() Corresponding author:

Rosie Saad

Tel. 1(514)340 4711 # 4273

Fax. 1(514)340 5913

e-mail:

Table of contents

1Model description

1.1Description of the model LANCA

1.2Illustrative example

2Land use impacts modeling

2.1Ecosystem quality curve for land use impacts modeling

2.2Characterization factors results for land occupation

2.2.1World Generic regionalization level

2.2.2Terrestrial Biomes regionalization level

2.2.3Holdridge Regions regionalization level

2.2.4Holdridge Life Zones regionalization level

2.3Characterization factors for land transformation

2.3.1Terrestrial Biomes regionalization level

3Assessment of spatial variability of characterization factors

1Model description

1.1Description of the model LANCA

The description of the methodology underlying the LANCA® calculation tool model used for the land use impact indicators is based on Baitz (2002) and Beck et al. (2010)[1]. Baitz’s approach was initially developed to quantify land use implications of industrial processes in LCA based on the concept of land functions. It provides a conceptual background and calculation instructions for different land use indicators calculated for specific land use cases within the LANCA® model.

Since the area used or transformed () and the occupation time () consist of inventory data for calculating the elementary flows, the LANCA® model can be used to calculate spatially differentiated characterization factors (CFs) for a range of land use activities, as done in the presented manuscript. Apart from the four indicators presented here, the LANCA® model also includes an indicator related to biotic production (also known as biomass production), which has been disregarded in this paper.

1. Erosion resistance:

This indicator measures the capacity of a land surface to resist water erosion and is based on physical soil characteristics and vulnerability. The model estimates erosion rate based on the universal soil loss equation (Wischmeier and Smith 1978). The calculation of the average annual loss in units of ton/(ha*year) is performed using several parameters, including soil texture, organic matter content, gravel content, summer precipitation and slope. This average naturally occurring rate is then corrected using a factor depending on the nature of the land use and the land cover (Bastian and Schreiber 1994). Such a correction factor is based on the degree of anthropogenic sealing of the land and reflects the influence of the vegetation cover to reduce erosion risk potential.

1. Physicochemical filtration:

The physicochemical filtration is one impact indicator to assess the soil function related to purification, filtering and buffering. This indicator measures the soil's capacity to adsorb substances diluted by retaining and immobilizing free cations in the soil solution and is expressed in units of cmolc/kgsoil. The performance of such an adsorption capacity is mainly based on the cation exchange capacity (CEC) of soil. The latter depends on soil texture, organic matter content, clay content and soil alkalinity (Bastian and Schreiber 1994; Umweltministerium Baden-Württemberg 1995). In addition, a factor driven by the vegetation cover and the sealing level of the soil is used to correct the chemical filtration capacity depending on the type of use.

1. Mechanical filtration:

The mechanical filtration is another impact indicator related to soil function of purification, filtering and buffering capacity. It measures the mechanical infiltration capacity in soil’s profile, and more specifically the rate of water passing per unit time (cm/day)(Schachtschabel et al. 1992). Such a flow represents the soil permeability based on two parameters; soil texture and the distance from the surface to groundwater level. Conversely to very fine texture (eg. clay), soil with a coarser texture (eg. sand) is characterized with higher soil permeability and consequently a better performance of the filter function. Moreover, the greater the distance between the surface and the water is, the longer the filtering distance and the performance of the filter is improved (Leser and Klink 1988; Baitz 2002). Finally, since vegetation cover protects the soil and promotes infiltration, a correction factor considering the anthropogenic sealing of the land surface is used to correct the influence of the type of use and land cover type.

1. Groundwater recharge[2]:

This indicator measures the soil's capacity to recharge groundwater resource based on the type of vegetation cover, climate hydrological and topographical conditions as well as the particle size and the soil texture (Marks et al. 1989). Based on the hydrological system regime, a water balance is used to yield natural groundwater recharge on a yearly basis (mm/year), evaluating the difference between the main inflows and outflows. Such flows consist of precipitation rates, evapotranspiration rates, soil field capacity as well as combination of the runoff coefficient. The latter depends on the slope and the distance between the surface and groundwater (Bastian and Schreiber 1994; Pfleiderer 1998). A correction factor considering the anthropization level of the land surface and the type of vegetation cover is used to correct the recharge rate calculated previously.

1.2Illustrative example

As an illustrative example and for guidance, a step by step procedure is shown considering one particular Holdridge Life zone, the Boreal moist forest. The input parameters values specific to this biogeographic unit and which were used for calibrating the LANCA® model Beck et al. (2010) are presented in the following Table S1.1.

Table S1.1 Spatially resolved average value of input parameters for the Boreal moist forest Holdridge Life Zone

Input parameters / Average data for Boreal moist forest
Soil texture / Medium sandy loam
Organic matter content (%) / 11
Gravel content (%) / 6
Cation exchange capacity (CEC) (cmolc/kgsoil) / 25.8
pH / 5.7
Depth to groundwater (m) / 3
Annual precipitation rate (mm/yearr) / 476
Annual evapo-transpiration rate (mm/yearr) / 320
Slope (°) / 3.5

The results from a permanent and annual crops land use type are obtained as following for each indicator. For additional information, a complete description of the model and the tables to which this section refers to can be found in Beck et al. (2010).

1. Erosion resistance:

Step 1: Erosion resistance class identification

–Based on the main parameter in modeling erosion resistance, the soil texture which is identified as being medium sandy loam is used for defining the corresponding erosion resistance class and which in this case is 3.1 (refer to table 2-1 in the model documentation).

–The erosion resistance class is further adjusted based on the organic content and the skeleton content (refer to table 2-2 in the model documentation). In the specific case of the Boreal moist forest data, the erosion resistance class does not need to be adjusted and still refers to the class 3.1.

Step 2: Average natural soil erosion estimate

–The average natural soil erosion (ANSE) is then determined based on the erosion resistance class and two additional parameters: the slope and the annual precipitation rate (refer to table 2-3 in the model documentation). Considering the specific data of the Boreal moist forest, the average natural soil erosion corresponds to 1.8 ton/(ha*year).

Step 3: Land use soil erosion rate calculation

–Soil erosion is strongly dependent on the type of land use. Thus, depending on the land use type modeled, a correction factor () is determined to adjust the average natural soil erosion () rate estimate previously calculated (refer to table 2-4 in the model documentation). The correction factor for permanent crops land use type is equal to 6. Therefore, the land use soil erosion () rate is obtained as a result of the average natural soil erosion rate estimate multiplied by the correction factor as per the following equation :

,

The value of refers to and results in a 10.8 ton/(ha.year) of soil eroded for permanent and annual crops in the Boreal moist forest.

Step 4: Soil erosion rate calculation for baseline reference

–The same procedure is followed to determine and which results in a value of 0.9 ton/(ha.year) of soil eroded.

Step 5: Characterization factor calculation

–Both and are used to determine CFs as per Eq (1) and Eq (2) in the manuscript. In order to reflect the resistance to soil eroded for this potential impact indicator, the quality difference is multiplied by -1.

–Consequently (-1) x (0.9 – 10.8) = 9.9 ton/(ha.year)

1. Physicochemical filtration:

Step 1: Determining cation exchange capacity

–Cation exchange capacity (CEC) is the main parameter to determine the soil physicochemical filtration capacity. This data is site-specific and needed as input in the model. In the specific case of the Boreal moist forest CEC is equal to 25.8 cmolc/kgsoil. This value refers to the soil natural physicochemical filtration capacity.

Step 2: Land use physicochemical filtration capacity calculation

–Physicochemical filtration capacity is dependent on the type of land use which is mainly driven by the anthropization level of the land surface. Thus, depending on the land use type modeled, a correction factor () is determined to adjust the average natural physicochemical filtration capacity () estimated previously (refer to table 4-5 in the model documentation).

–The correction factor for permanent crops land use type is equal to 0.05. Therefore, the land use physicochemical filtration capacity () rate is obtained as a result of the average natural physicochemical filtration capacity estimate multiplied by 1 minus the correction factor as per the following equation :

The value of refers to

Step 3: Land use physicochemical filtration capacity for baseline reference

–The same procedure is followed to determine which refers to the average natural physicochemical filtration capacity. This result corresponds to 25.8cmolc/kgsoil.

Step 5: Characterization factor calculation

–Both and are used to determine CFs as per Eq (1) and Eq (2) in the manuscript.

–Consequently (25.8 –24.5) = 1.29 cmolc/kgsoil

1. Mechanical filtration:

Step1: Soil texture class identification

–Based on the soil texture identified for a given biogeographic unit, a soil texture class is identified (refer to figure 2-2 in the model documentation). In the specific case of the Boreal moist forest, the soil texture class identified is VI.

Step2: Determining water permeability

–Then, based on the soil texture class identified the water permeability of the soil is determined (refer to table 2-5 in the model documentation). Thus two parameters are identified, the permeability group and the filtration capacity being of 3 and 25 cm/day respectively for the Boreal moist forest.

–Since soil filtration capacity is influenced by the filtration distance, i.e. the distance from surface to groundwater level. Indeed the filtration capacity increases as the distance is longer. Thus, the water permeability is further adjusted when the distance is greater than 10 m or smaller than 0.8 m (refer to table 4-6 in the model documentation).

–The adjusted value, if needed, refers to the natural mechanical filtration capacity. In this specific case, it corresponds to 25 cm/day.

Step 3: Land use mechanical filtration capacity calculation

–Mechanical filtration capacity is dependent on the type of land use which is mainly driven by the anthropization level of the land surface. Thus, depending on the land use type modeled, a correction factor () is determined to adjust the average natural mechanical filtration capacity () estimated previously (refer to table 4-5 in the model documentation).

–The correction factor for permanent crops land use type is equal to 0.05. Therefore, the land use mechanical filtration capacity () rate is obtained as a result of the average natural mechanical filtration capacity estimate multiplied by 1 minus the correction factor as per the following equation :

The value of refers to

Step 4: Land use groundwater recharge rate for baseline reference

–The same procedure is followed to determine and which results in a value of 25 cm/day.

Step 5: Characterization factor calculation

–Both and are used to determine CFs as per Eq (1) and Eq (2) in the manuscript.

–Consequently (25 –23.75) = 1.25 cm/day

1. Groundwater recharge:

Step 1: Determining the field capacity

–The first parameter used in the modeling is the soil texture which is identified as being medium sandy loam. The latter determines the corresponding field capacity class and which in this case is 4 (refer to figure 2-4 and table 4-10 in the model documentation).

–From the previously determined field capacity, a field capacity mean value (in mm) is further (refer to table 2-7 in the model documentation).

Step 2: Estimate of groundwater recharge rate

–For very specific cases of land use and parameters data range for field capacity, annual precipitation and annual evapotranspiration, Eqs.(3) to (6) in the model documentation should be used.

–If the data ranges of these parameters are not applicable, then a water balance is used to yield an estimate of a natural groundwater recharge () on a yearly basis (mm/year). The latter is a non-corrected groundwater recharge rate and is calculated as the difference between annual precipitation () and annual evapotranspiration () as per:

= 476 – 320 = 156 mm/year

–For very specific cases of land use and parameters data range for field capacity, annual precipitation and annual evapotranspiration, Eqs. (3) to (6) in the model documentation should be used.

–In order to account for surface runoff, the groundwater recharge rate is corrected and is divided by the runoff coefficient () determined from the hydromorphology class and the slope (refer to table 2-8 in the model documentation). Based on the data for the Boreal moist forest, the runoff coefficient is of 1.5

Step 3: Land use groundwater recharge rate calculation

–Groundwater recharge rate is dependent on the type of land use which is mainly driven by the anthropization level of the land surface. Thus, depending on the land use type modeled, a correction factor () is determined to adjust the average groundwater recharge rate () rate estimate previously calculated (refer to table 4-5 in the model documentation).

–The correction factor for permanent crops land use type is equal to 0.05. Therefore, the land use groundwater recharge () rate is obtained as a result of the average natural groundwater recharge rate estimate multiplied by 1 minus the correction factor as per the following equation :

The value of refers to

Step 4: Land use groundwater recharge rate for baseline reference

–The same procedure is followed to determine and which results in a value of 104 mm/year.

Step 5: Characterization factor calculation

–Both and are used to determine CFs as per Eq. (1) and Eq. (2) in the manuscript.

–Consequently (104 – 99) = 5 mm/year

2Land use impacts modeling

2.1Ecosystem quality curve for land use impacts modeling

The ecosystem quality curve is used to calculate the magnitude of land use impacts.

Fig S2.1 Ecosystem quality curve indicating transformation (dark gray) and occupation (light gray) impacts magnitudes; adapted from Lindeijer (2000), Milà i Canals et al. (2007a) and Koellner et al. (2103)

Where:

A is the occupied or transformed area;
t2 to t3 corresponds to the duration of the occupation stage ( );
t3 to t4 corresponds to the regeneration time ( ) needed for the ecosystem to reach a relaxed stated considered as a quasi-natural state;
and correspond to the ecosystem quality for the baseline reference () reaching a relaxed state (i.e. a quasi naturel state) and for the land use type () respectively.

2.2Characterization factors results for land occupation

The results of CFs estimates for occupation developed for all three global spatial resolution scale (nine Holdridge regions, thirty-eight Holdridge Life zones and fourteen terrestrial biomes level) and the non-spatial one (World average) are shown in Tables S2.1 to S2.16. Results noted with N/A indicate no vegetative soil development on the corresponding biogeographic unit (Polar Regions). In addition, negative values of CF mean an increase in the respective ecosystem service and positive values express the reduction of a service being a negative ecological impact.

2.2.1World Generic regionalization level

Table S2.1 Characterization factors for land occupation for the impact indicator erosion resistance using the World Generic non spatial regionalization level

Characterization Factor: / ∆ Erosion resistance capacity (ton/(ha.year))
Global non spatial / Land use type
Forest
(1) / Shrubland (3) / Grassland (4.1) / Pasture/
Meadow (4.1) / Permanent and annual crops (5) / Urban (7.1) / Urban, green areas
(7.1.4)
World Generic / -1.43 / 1.53 / -1.43 / 13.37 / 31.13 / 79.42 / 13.37

Table S2.2 Characterization factors for land occupation for the impact indicator physicochemical filtration using the World Generic non spatial regionalization level

Characterization Factor: / ∆ Physicochemical filtration capacity (cmolc/kgsoil)
Global non spatial / Land use type
Forest
(1) / Shrubland (3) / Grassland (4.1) / Pasture/
Meadow (4.1) / Permanent and annual crops (5) / Urban (7.1) / Urban, green areas (7.1.4)
World Generic / 0.82 / 0.00 / 0.00 / 0.82 / 0.82 / 15.56 / 3.28

Table S2.3 Characterization factors for land occupation for the impact indicator mechanical filtration using the World Generic non spatial regionalization level

Characterization Factor: / ∆ Mechanical filtration capacity (cm/day)
Global non spatial / Land use type
Forest
(1) / Shrubland (3) / Grassland (4.1) / Pasture/Meadow (4.1) / Permanent and annual crops (5) / Urban (7.1) / Urban, green areas (7.1.4)
World Generic / 1.74 / 0.00 / 0.00 / 1.74 / 1.74 / 33.09 / 6.97

Table S2.4 Characterization factors for land occupation for the impact indicator groundwater recharge rate using the World Generic non spatial regionalization level

Characterization Factor: / ∆ Groundwater recharge rate capacity (mm/year)
Global non spatial / Land use type
Forest
(1) / Shrubland (3) / Grassland (4.1) / Pasture/
Meadow (4.1) / Permanent and annual crops (5) / Urban (7.1) / Urban, green areas (7.1.4)
World Generic / 1.8 / 19 / 17 / 24 / 24 / 154 / 46

2.2.2Terrestrial Biomes regionalization level

Table S2.5 Characterization factors for land occupation for the impact indicator erosion resistance using the Terrestrial Biomes regionalization level