Land Use Damage Assessment and its Application on Resource Extraction and Waste Landfill Impact Categories in LIME2

Ryouta Ii, Kazuko Yamaguchi, Ai Okada*, Norihiro Itsubo**, Atsushi Inaba***

Pacific Consultants Co., Ltd. Shinjuku-daiichiseimei-bldg., 2-7-1, Nishi-shinjuku, Shinjuku-ku, Tokyo, 163-0730, Japan

Ryouta.Ii@tk.pacific.co.jp *Civil Engineering Center, Tokyo Metropolitan Government

1-9-15 Shinsuna, Koto-ku, Tokyo, 136-0075, Japan **Musashi Institute of Technology, 3-3-1 Ushikubo-nishi, Tuzuki-ku, Yokohama, Kanagawa, 224-0015, Japan

***University of Tokyo, 5-1-5, Kashiwanoha, Kashiwa, Chiba, 277-8568, Japan

Keywords: uncertainty analysis, LCIA, land use, NPP, extinction risk

ABSTRACT Land use is one of the impact categories covered in LIME (Life-cycle Impact assessment Method based on

Endpoint modeling) version 1 from the beginning of its completion in FY 2002. The damage assessment method on ecosystem was developed not only for land use but also for its related impact categories in a typical product life-cycle, namely, resource extraction and waste landfill. The damage indicators are the increase of extinction risk of plant species and the loss of net primary production and they correspond to the damage categories of biotic natural environment and biotic natural resources in the LIME framework, respectively. In the development of the second version of LIME (LIME 2) since 2003, we executed quantitative uncertainty assessments on the damage functions of these categories by Monte Carlo approach. The result was compiled to the list of updated damage factors with their statistics values and information of the probability distribution functions. Furthermore, it helped to examine the future direction of the improvement of our damage analysis as well as their characteristics and limitations. In this presentation, the achievements and problems about the land use damage assessment and its application on resource extraction and waste landfill impact categories in LIME2 will be introduced and compared with some of the other methods in Europe.

LAND USE IMPACT IN LIME FRAMEWORK LIME (Life-cycle Impact assessment Method based on Endpoint modeling) is an LCIA method developed for Japan, and the project to develop its second version, LIME2, had been launched in 2002. The major progresses since LIME1 are as follows [1]: ・ Introduction of 2 new impact categories (transport

noise and indoor air pollution) ・ Application of uncertainty analysis on the damage

factors, weighting factors and integrating factors by the Monte Carlo approach

・ Improvement of representativeness and credibility of weighting factors for integration (based on a panel survey on a some thousand people)

Fig 1 shows the basic framework of LIME2. The safeguard subjects for ecosystem are biodiversity and primary production. The damage indicator of biodiversity is the increase of extinction risk of plant species and that of primary production is the amount of lost NPP (net primary production).

Land use is one of the important impact categories for ecosystem safegurad subjects in LIME. In addition, the damage factors are provided for resource extraction and waste landfill by applying the damage assessment of land use.

LAND USE DAMAGE ASSESSMENT IN LIME

In LIME, the damage function of land use has been developed for NPP and biodiversity[2, 3]. Damage Assessment for NPP (Primary Production) The aspect of land use in the damage function for NPP is classified into the continuation of land use (land occupation) and land transformation (land change).

The damage of land occupation is calculated assuming that potential NPP is not fully achieved while the land is occupied. The damage due to land transformation is calculated by the loss of NPP during the period required for recovery of vegetation from altered to its potential state. [Damage factor of land occupation] (Fig. 2-1) ( ) ap

OccNPP NPPNPPaDF −= (kg/m2/year) (1)

[Damage factor of land transformation] (Fig. 2-2) ( ) ( ) ( )pa

OccNPPpa

TransNPP TaDFabDF >−>− ×= α,

( ) ( )pbOccNPPpb TbDF >−>− ×−α (kg/m2) (2)

b,a: Category of vegetation or landuse (before, after) T: Time required for recovery of land productivity α: 0.5 (Recovery of NPP is assumed to progress linearly.)

5

Fig.1 Framework of LIME2

The damage factors are prepared for each of 30

types of vegetation and 9 of land-use categories. Damage Assessment for Biodiversity In the damage function of land use for biodiversity, the damage indicator is the sum of increases in inverse number of expected time to extinct (T) of endangered vascular plant species due to land transformation for a unit of land. The indicator is called ‘EINES’ (Expected Increase in Number of Extinct Species) and calculated as follows:

( )∑∑ −=Δ=s ,a,s

)/1()/1()/1( ssbs TTTEINES (3)

The calculation is based on the assessment method developed by Matsuda et al.[4] following the steps shown in Fig. 3 below. The major data sources are the Red Data Book (RDB)[5] and Environmental Impact Statements (EIS's) prepared for the projects involving large-scale land alteration. The damage factors are prepared for roads, landfills, gravel pits and railways.

Resource Extraction and Waste Landfill At the time LIME project had started, land use activities related to resource extraction and waste landfill

Fig.3 The flow of calculating damage intensity per unit area of land changed

RDB (Red Data Book): ranks of population anddecreasing rates of endangered species

EIS (Environmental Impact Statements):populations in the target area

Tbefore: T without the land change Tafter:T with the land change

observed populations in the EISstatistical models of national population andaverage degreasing rates

Statistical model of expected time to extinct (T)

Increased extinction risk for the target area IER=

Increased extintion risk per unit area UAR=IER / size of target area

Average damage intensity per unit area for rock extraction in Japan

if lost by the landchange

damage assessment for one project (one EIS) of resource extraction

regression from results of

simulation m

odels

∑∑ −=Δ )/1/1()/1( beforeafter TTT

Average damage intensity per unit area for land use change in Japan

Damage assessment for one project (one EIS) of land use change

statistical models of national population and average decreasing rates

t-1:before transformation, t0:after transformation, t1: abandonment of land use a, t2: time when NPP recovers to the former land use level (NPPb), t3:time whose NPP recovers to the potential level (NPPp).

Fig. 2-1 Damage assessment Fig. 2-2 Damage assessment (land occupation) (land transformation)

Fig.4 Calculation flow of damage factors for resource extraction

had been rarely included in the inventory of LCA studies. Therefore, the damage factors for these activities were established by identifying the area and utilization time of land required for these activities and relating them to the damage factors for landuse.

For the activities that take place in Japan, the landuse inventory was prepared by referring to the EIS’s, the data of existing landfill facilities and other sources, and directly connected with the damage factors for landuse. For resource extraction in other countries, on the other hand, the area of landuse was estimated by a symple equation (eq.4) shown in Fig.4. The damage factors on NPP were obtained by estimating the global NPP, and those on biodiversity were set by correcting the Japanese damage factors. The calculation flow is shown in Fig. 4.

LIME2: REVISEMENT THROUGH UNCERTAINTY

ANALYSIS In the development of LIME2, uncertainty

analyses were conducted, as well as addition and update of foundamental data for parameter setting, to revise the damage factors[7,8]. The resulting statistics and fitted probability density distribution are provided as the list of damage factors. The evauated major parameters are shown in Table 1.

The results show that the uncertainty is higher in the damage factors on biodiversity than on NPP,

probably due to grater geographic variability; as EINES reflects the differences in extinction risks among species, they are influenced by the variations in plant species that grow in the area to be transformed. Another uncertainty is the correction of damage factors on resource extraction to apply Japanese landuse factors to other countries. Also, extraction of low-grade metal involve the variation in the estimated area to be converted, acting as an important factor of uncertainty. For landfill, the differences in the area used for each landfill are identified as a major uncertainty factor.

DISCUSSION

Comparison of Damage Indicators with Other LCIA Methods

The damage assesssments on biotic production or life support functions are targeting NPP or soil quality. Lindeijer [9], a pioneer project in Europe, had developed an assessment method for fNPP (free NPP), but assessed the land occupation and land change separately. LIME, on the other hand, presented a methodology in which land occuparion and land change are integrated [2]. Although Eco-indicator99 [10] attempts to integrate them, too, by applying the concept of restoration time, it does not take α in Equation (2) into account. Our formulation –including of α and assuming it to be 1/2– is rather similar to Köllner et al. [11].

ore

deposit

concentration

tailings

overburden

Conceptual figure of the calculation using the equation (4)

HgOm

DGCmA ii

iiii

11111×××+×××= (4)

A: size of land changed due to ore extration and stack of hidden flows (overburden and tailings) [m2] i: index number of resource (resource category; e.g. Fe) mi:weight of metal i (life-cycle inventory) [kg] Ci: ore grade Gi: density of ore [kg/m3] Di: thickness of deposit [m] Oi: raito of hidden flows to ore g: density of hidden flows [kg/m3] H: stack height of hidden flows [m]

the damage factors on biodiversity (EINES/kg)

m2/kg

(1) size of land transformation by 1kg resource extraction

(2) damage intensity on biodiversity by land transformation

1) damage intensity per unit area of land change due to resource extraction in Japan (EINES/m2)

2) conversion of the damage amount into that for the countries of exporters to Japan (EINES/m2)

EINES/m2

the damage factors on primary production (NPP-kg/kg) NPP/m2/yr

period

(3) damage intensity on NPP by land transformation(estimated by Chikugo Model [6])

National distribution, or at the specific grid (the mining site)

(example: metals & coal)

In Europe, damage assessment on biodiversity is often based upon SAR (Speicies-Area relationship) in community ecology (Goedkoep et al. [10], Lindeijer [9], Köllner[12], Köllner et al.[13], Schmidt [14]). PDF (Potentially Disappeared Fraction) is one of the typical damage indicators derived from SAR. PDF, however, depends on the different concept from PAF which is the major indicator in the damage asssessment of ecotoxic chemicals. Moreover, it is pointed that they do not reflect the relative scarcity of species [15].

The damage assessment of land use in LIME is based on the risk assessment method (extinction probability) in conservation ecology. As stated above, ‘EINES’ considers the differences in average lifetime among species. It is also applied in the ecotoxicological assessment. There are, however, some restrictions; that is, they evaluate land change but does not deal with land occupation and restoration time. Furthermore, applicability of foundamental data for assessment is more limited than in SAR, and, therefore, uncertainty in the assessment for other countries is high. Comparison of Land Use Interventions by Resource Extraction The area of land converted for extraction of a unit of some metals are compared with that in ecoinvent2 [16] (Table 2). The results show that the amount is larger in LIME2 than in ecoinvent2. The reasons may include that the landfill areas of talings in LIME2 are not included in the ecoinvent in the table (Cu); LIME2 does not distinguish underground mines and open mines (Pb); and the differences due to overburden (Fe). Further study is needed.

Table 2 Comparison of land transformation (m2/kg) LIME2 (median) Ecoinvent (ver.2.01)

Cu 5.14E-03 7.93E-05 – 3.33E-04 Pb 6.17E-04 5.20E-05 Fe 3.34E-05 4.60E-06

ACKNOWLEDGEMENT

This research was funded by a grant from the New Energy and Industrial Technology Development Organization of Japan (NEDO).

REFERENCES [1] Itsubo N. and Inaba A.; Development of LIME2 -Updated

Comprehensive Japanese LCIA Methodology-, Proc. of the 7th International Conference on EcoBalance, Tsukuba, pp.165-166, (2006)

[2] Nakagawa A. et. al.: Development of Life-Cycle Impact Assessment Method for Land use Based on Endpoint Damage. Proc. of the 5th Int’l. Conference on EcoBalance, (2002)

[3] Ii R. et al.: Life Cycle Impact Assessment for Office Building Considering Land Use and Related Impact Categories, Proc. of the 5th International Conference on EcoBalance, (2002)

[4] Matsuda H. et. al. Assessment of the impact of the Japanese 2005 World Exposition project on the extinction risk of vascular plants. Chemosphere, 53(4), 325-336, (2003)

[5] Compiled by Environment Agency: Red Data Book 2nd ed. – Volume 8, Vascular Plants, (2000) (in Japanese)

[6] Uchijima Z. and Seino H.: Agroclimatic Evaluation of Net Primary Productivity of Natural Vegetations. J.Agr.Met., 40(4), 343-352, (1985)

[7] Yamaguchi K. et al.: Uncertainty Analysis of Ecosystem Damage Function Caused by Land Use, Prof of the 7th International Conference on EcoBalance, Tsukuba, (2006)

[8] Ii R. et al.: Uncertainty Analysis of Ecosystem Damage Function caused by Resource Extraction, Prof of the 7th International Conference on EcoBalance, Tsukuba, (2006)

[9] Lindeijer E.: Biodiversity and life support impacts of land use in LCA. Journal of Cleaner Production 8, pp.313-319, (2000).

[10] Goedkoop, M, Spriensma, R, “The Eco-indicator 99, A damage oriented method for life cycle impact assessment”, Pré consultants, (1999)

[11] Köllner T., Scholz R.W.: Assessment of Land Use Impacts on the Natural Environment - Part 1. Int J LCA, 12(1), pp.16-32, (2007)

[12] Köllner T.: Species-pool efect potentials (SPEP)as a yardstick to evaluate land-use impacts on biodiversity. J Clean Prod, 8, pp.293-311, (2000)

[13] Köllner T., Scholz R.W.: Assessment of Land Use Impacts on the Natural Environment - Part 2. Int J LCA, 13(1), pp.32-48, (2008)

[14] Schmidt J H: Development of LCIA characterisation factors for land use impacts on biodiversity, J Clean Prod., 16(18), pp. 1929-1942, (2008)

[15] Milà i Canals et al.: Key elements in a framework for land use impact assessment in LCA. Int J LCA ,12(1), pp.5-15, (2007)

[16] ecoinvent2, http://www.ecoinvent.ch/

Table 1 Major parameters in uncertainty analysis Impact Category Module Evaluated Parameters (Probability density distributions were set for these.)

NPP

Parameters of Chikugo model for potential NPP, current NPP for each vegetation type and geographical distributions of 2 types of NPP; recovery time

Land Use

EINES

Parameters of regression models for estimating expected time to extinction of the plant species, Estimated population of the species in Japan and in the surveyed area in the EISs, variation of the EINES/m2 among the evaluated EISs

Land use Parameters of estimating the size of land changed for the extraction (e.g. ore grade, hidden flows, thickness of deposits) for each metal, mining duration

NPP World NPP distribution and unknown location of extraction, recovery time

Resource extraction (e.g. metals)

EINES Density of # of endangered species in each country, parameters of regression model for EINES conversion from Japan to the other countries, probability density function of the EINES factor

Land use Density for each waste type, variation among landfills, parameters for estimating landfill volume and area conversion ratio, duration until abandonment

Waste landfill

NPP&EINES (see ‘Land use’ impact category)