5.1 Treatment of irrigation requirements in water footprint (WF) calculations
Methods by which irrigation requirements are treated may make a difference to water footprint values. The blue water footprint (WFblue) in the study of Chapagain and Hoekstra (2007) was calculated based on the presumption that irrigation demand could be completely satisfied, while that in our thesis was calculated based on real irrigation volumes of sampled tea farms.
An identical green-and-blue water footprint of PingLin tea, being 16.424 cubic metres per kilogram, were calculated whether by adhering to the method of Chapagain and Hoekstra (2007) or that of our thesis since the irrigation requirements of PingLin were zero throughout the year on a growth period basis.
Different green-and-blue water footprints calculation methods result in different WF values in terms of MingJian tea, as evidenced by the fact that the annual green-and-blue water footprints of MingJian tea were 7.137 and 6.798 cubic metres per kilogram based on the method of Chapagain and Hoekstra (2007) and that of our thesis respectively. Thus, the annual green-and-blue water footprint of MingJian tea calculated based on the method of Chapagain and Hoekstra (2007) was 5% greater than that calculated based on the methods of our thesis. The 5%
difference is due to low availability of irrigation water on BaGua tableland, on which the sampled tea farms of MingJian are located.
In conclusion, an identical green-and-blue water footprint of PingLin tea was obtained whether following the methods of our thesis or the that of Chapagain and Hoekstra (2007). Meanwhile, the green-and-blue water footprint of MingJian tea calculated based on the method of Chapagain and Hoekstra (2007) was slightly greater than that calculated based on the methods of our thesis.
5.2 Treatment of product fractions (PF) in water footprint (WF) calculations
Product fractions of tea (PF), which is the ratios of dried tea leaves to fresh ones, vary between our thesis and the study of Chapagain and Hoekstra (2007). According to the field surveys of either PingLin or MingJian, the PF is measured as 0.2 regardless of harvesting methods. Meanwhile, in the study of Chapagain and Hoekstra (2007), the PF is presumed to be 0.26 universally. Thus, PF value of our field survey is 77% of that used in the study of Chapagain and Hoekstra (2007). In conclusion, in the study of Chapagain and Hoekstra (2007), more accurate results
52
may be obtained if the difference in pf between each tea producers is considered.
5.3 Virtual water over spaces (VWS) and water use intensity
Virtual water over spaces (VWS) is the medium by which water use intensity and the environmental burden created by an industrial activity are evaluated.
Regardless of the level of VWS, greater VWS implies higher water use intensity and thus higher environmental burden an industrial activity imposes on the environment.
Therefore, VWS correlates positively with corresponding water use intensity and the environmental burden of an industrial activity.
At regional and annual level, the fact that VWS of PingLin is 5.8 times as much as that of MingJian suggests that water use intensity tea production in PingLin is higher than that of MingJian on an annual basis (Table 5.3.1). Thus, at regional and annual level, tea production of PingLin imposes greater burden on the environment in relation to that of MingJian.
Table 5.3.1 VWStotal and WFtotal of PingLin and MingJian Title
Growth Period
VWStotal (mm/ /day) WFtotal ( )
PingLin MingJian PingLin MingJian
Spring 70.826 6.337 334.873 36.96494 vary with growth periods, as evidenced by following facts.
First, in a spring growth period and at regional level, the fact that the VWS of PingLin is 11.2 times as much as that of PingLin suggests that water use intensity of PingLin is higher than that of MingJian in a spring growth period (Table 5.3.1). Thus, in a spring growth period and at regional level , tea production of PingLin imposes greater burden on the environment in relation to that of MingJian.
Second, in a summer growth period and at regional level, the fact that the VWS of MingJian is 2.4 times as much as that of PingLin suggests that water use intensity of MingJian is higher than that of PingLin in a summer growth period (Table 5.3.1).
Thus, in a summer growth period and at regional level, tea production of MingJian
53
imposes more burdens on the environment in relation to that of PingLin.
Third, in a fall growth period and at regional level, the fact that the VWS of PingLin is 3 times as much as that of MingJian suggests that water use intensity of PingLin is higher than that of MingJian in a fall growth period (Table 5.3.1). Thus, in a fall growth period and at regional level, tea production of PingLin impose greater burdens on the environment in relation to that of MingJian.
Fourth, in a winter growth period and at regional level, the fact that the VWS of PingLin is 5.7 times as much as that of MingJian suggests that water use intensity of PingLin is higher than that MingJian in a winter growth period (Table 5.3.1). Thus, in a winter growth period and at regional level, tea production of PingLin imposes more burdens on the environment in relation to that MingJian.
In conclusion, at regional and annual level, water use intensity of tea production of PingLin is higher than that of MingJian, so that tea production of PingLin imposes greater burdens on the environment than that of MingJian.
Meanwhile, at regional and growth period level, water use intensity of tea production of PingLin is higher than that of MingJian in all growth period except for a summer one, so that tea production of PingLin imposes greater burdens on the environment than that of MingJian in all growth periods except for a summer one.
5.4 Water footprint (WF) and water use efficiency
Water footprint (WF) is the medium by which water use efficiency of an industrial activity and the resulting environmental burdens created by consuming the products of that industrial activity are evaluated. Regardless of the level of WF, in terms of a product, greater WF implies lower water use efficiency for its producers and greater virtual water consumption for its consumers. Moreover, higher virtual water consumption further implies greater burdens a consumer exhausting that product imposes on the environment.
At regional and annual level, the fact that the WF of PingLin tea is 13.3 times as much as that of PingLin tea suggests that water use efficiency of tea production in MingJian is higher than that in PingLin on an annual basis. Thus, in relation to MingJian tea, consumers impose greater burden on the environment by drinking PingLin tea according to regional-and-annum based WF of made tea and according to identical amount of tea consumption.
At regional and growth period level, difference between PingLin and MingJian in terms of water use efficiency of tea productions and the environmental burdens created by tea consumers vary with growth periods, as evidenced by the following speculations.
54
First, in a spring growth period and at regional level, the fact that the WF of PingLin tea was 9.1 times as much as that of MingJian tea implies that water use efficiency of tea production in MingJian is higher than that in PingLin. Moreover, in relation to MingJian spring tea, consumers impose greater burden on the environment by drinking PingLin spring tea based on the WF of PingLin and MingJian throughout a spring growth period and based on identical amount of tea consumption.
Second, in a summer growth period at regional level, the fact that WF of PingLin tea is 15.7 times as much as that of MingJian tea implies that water use efficiency of tea production in MingJian is higher than that in PingLin. Moreover, in relation to MingJian summer tea, consumers impose greater burden on the environment by drinking PingLin summer tea based on the WF of PingLin and MingJian tea throughout a summer growth period and based on identical amount of tea consumption.
Third, in a fall growth period at regional level, the fact that the WF of PingLin tea is 6 times as much as that of MingJian tea implies that the water use efficiency of tea production in MingJian is higher than that in PingLin. Moreover, in relation to MingJian fall tea, consumers impose greater burden on the environment by drinking PingLin fall tea based on the WF of PingLin and MingJian tea throughout a fall growth period and based on identical amount of tea consumption.
Fourth, in a winter growth period at regional level, the fact that the WF of PingLin tea was 7.2 times as much as that of MingJian tea implies that water use efficiency of tea production in MingJian is higher than that in PingLin. Moreover, in relation to MingJian winter tea, consumers impose greater burdens on the environment by drinking PingLin winter tea based on the WF of PingLin and MingJian throughout a winter growth period and based on identical amount of tea consumption.
In conclusion, water use efficiency of tea production is higher in MingJian than in PingLin whether at regional-and-annual level or at regional-and-growth-period level. Thus, whether at regional-and-annual level, or at regional-and-growth-period level, a consumer imposes greater burdens on the environment by drinking PingLin tea in relation to drinking MingJian tea.
5.5 Appropriateness evaluation of tea production in PingLin and MingJian from a global perspective
Appropriateness of tea production in PingLin and MingJian from a global perspective is evaluated by comparing the WF of either PingLin or MingJian tea
55
with those of other tea producers worldwide, whose information is available from the study of Chapagain and Hoekstra (2007). Higher WF implies lower water use efficiency and thus lower appropriateness of tea production, while lower WF implies higher water use efficiency and thus higher appropriateness of tea production.
In evaluation of appropriateness of tea production, only green-and-blue water footprints are considered since grey water footprints have been neglected in the study of Chapagain and Hoekstra (2007), which provides WF of thirteen major tea producers around the world.
At global scale, water use efficiency and appropriateness of tea production varies between PingLin and MingJian. The fact that the water footprint of PingLin tea ranks the second highest among 13 major tea producers implies that water use efficiency of PingLin is rather low from a global perspective. Meanwhile, the fact that the water footprint of tea produced in MingJian ranks the third lowest among 13 major tea producers implies that water use efficiency of MingJian is rather high from a global perspective.
In conclusion, on global scales, based on the water use efficiency of tea production among 13 major tea producers worldwide, tea production of PingLin is less appropriate due to its rather lower water use efficiency. Meanwhile, tea production of MingJian is rather appropriate due to its higher water use efficiency.
56
Table 5.5 Green-and-blue-water-footprints of major tea producers Countries Water footprint of made tea
( /kg)
China 16.604
PingLin 16.424*
Uganda 15.61
Tanzania 13.377
Indonesia 12.395
Sri Lanka 12.247
South Africa 10.965
Argentina 9.208
Brazil 8.41
Mauritius 7.191
Turkey 7.053
Japan 6.95
MingJian 6.798 a
India 4.978
Bangladesh 1.305
Source: Chapagain and Hoekstra, 2007
a The green-and-blue water footprints of PingLin and MingJian are obtained from this thesis.
5.6 Factors influencing green virtual water over spaces of tea gardens (VWSgreen) The presumption that daily average potential evapotranspiration (PET) is a factor influencing green virtual water of tea garden (VWSgreen) is supported by correlations between VWSgreen and daily average PET at various temporal and geographical level.
At regional and annual level, VWSgreen correlates positively and very strongly with corresponding daily average PET with the correlation being 1. Thus, at regional and annual level, within PingLin and MingJian, the very strong correlation between VWSgreen and corresponding daily average PET implies that daily average PET may play a part in determining the region-and-annum based VWSgreen.
At regional and growth period level, the VWSgreen correlates positively with corresponding daily average PET to different degrees, depending on investigated regions selected. Within PingLin, at regional and growth period level, VWSgreen correlates positively and very strongly with corresponding daily average PET, with
57
the correlation coefficient being 1. Within MingJian, at regional and growth period level, the VWSgreen correlates positively and strongly with corresponding daily average PET with the correlation coefficient being 0.741. Thus, at regional and growth period level, within either PingLin or MingJian, the strong to very strong correlations between VWSgreen and corresponding daily average PET imply that daily average PET is a possible factor influencing region-and-growth-period based VWSgreen.
Table 5.6 VWSgreen[i,j] and daily average PET Region
Growth period
PingLin PingLin MingJian MingJian
VWSgreen[i,j] PET (mm) are calculated by dividing PET of each growth period by the length of a corresponding growth period.
In conclusion, whether at regional and growth period level or at regional and annual level, within either PingLin or MingJian , VWSgreen is possibly affected by daily average PET based on strong to very strong correlation between VWSgreen and daily average.
5.7 Factors influencing grey virtual water over spaces of tea gardens (VWSgrey)
The presumption that nitrogen input per unit area (NI/A) is a factor influencing grey virtual water of tea gardens (VWSgrey) is supported by correlations between VWSgrey and NI/A at various temporal and geographical level.
At regional and annual level, within either PingLin or MingJian, the VWSgrey correlates positively and very strongly with corresponding nitrogen input per unit area (Table 5.7.1; Table 5.7.2), with the correlation coefficient being 1. Thus, at regional and annual level, within either PingLin or MingJian, the very strong positive correlation between VWSgrey and corresponding NI/A suggests that NI/A
58
possibly plays a role in determining the region-and-annum based VWSgrey.
At regional and growth period level, VWSgrey correlates positively with corresponding NI/A to different degrees, depending on the investigated regions selected. Within PingLin, at regional and growth period level, VWSgrey correlates positively and very strongly with corresponding NI/A with the correlation coefficient being 0.994 (Table 5.7.1; Table 5.7.2). Within MingJian, at regional and growth period level, VWSgrey correlates positively and strongly with corresponding NI/A with the correlation coefficient being 0.741 (Table 5.7.1; Table 5.7.2). Thus, at regional and growth period level, within either PingLin or MingJian, the strong to very strong positive correlations between VWSgrey and corresponding NI/A imply that the region-and-growth-period based VWSgrey may be affected by NI/A .
At farm and annual level, within either PingLin or MingJian, or within PingLin along with MingJian, the VWSgrey correlates positively and very strongly with corresponding NI/A, with all the correlation coefficients being 1 (Table 5.7.1; Table 5.7.2). Thus, at individual farm and annual level, within either PingLin or MingJian, or within PingLin along with MingJian, the very strong positive correlations between the VWSgrey and corresponding NI/A suggest that the farm-and-annum based VWSgrey is possibly affected by NI/A.
In conclusion, within either PingLin or MingJian, or within PingLin along with MingJian, at various temporal and geographical level, the presumption that VWSgrey is possibly affected by NI/A is tenable based on the strong to very strong correlations between VWSgrey and NI/A.
Table 5.7.1 Nitrogen input per unit area (NI/A) by farm within PingLin (kg/ ) Farm title
Growth period
Farm A Farm B Farm C Farm D Farm E NI/A[i,j]
Spring growth period 0.063 0.003 0.063 0.022 1.979 0.118 Summer growth period 0.000 0.001 0.000 0.000 0.000 0.000 Fall growth period 0.000 0.001 0.024 0.011 0.000 0.012 Winter growth period 0.031 0.000 0.008 0.011 0.000 0.014 Throughout 2011 0.094 0.005 0.095 0.045 1.979 0.143
Note. Farm A refers to Tian Hsian; Farm B refers to Wen Ping; Farm C refers to Jian
Yuan; Farm D refers to Kao Yi Fa; Farm E refers to Tian Shian.
59
Table 5.7.2 Nitrogen input per unit area (NI/A) by farm within MingJian (kg/ ) Farm title
Growth period Farm F Farm G Farm H Farm I Farm J NIA[i,j]
Spring growth period 0.006 0.003 0.013 0.008 0.003 0.006 Summer growth period 0.004 0.008 0.008 0.006 0.002 0.005 Fall growth period 0.003 0.000 0.005 0.004 0.001 0.003 Winter growth period 0.002 0.005 0.003 0.002 0.001 0.002 Throughout 2011 0.015 0.015 0.029 0.020 0.007 0.016
Note. Farm F refers to Yuan Chih; Farm G refers to Jia Tian; Farm H refers to Yuan
Bei; Farm I refers to Wu Cha Nung Hu; Farm J refers to Lui Pin.
5.8 Factors influencing green water footprints (WFgreen) of made tea
The presumption that yield per unit area of tea gardens (Y/A) is a possible factor influencing green water footprint of made tea (WFgreen) is supported by different degrees of correlations between WFgreen and Y/A at various temporal and geographical level.
At regional and annual level, within both PingLin and MingJian, the WFgreen correlates negatively and very strongly with corresponding annual Y/A, with the correlation coefficient being -1. Thus, at regional and annual level, within both PingLin and MingJian, the very strong negative correlation between WFgreen and the corresponding annual Y/A suggests that the annual Y/A is a possible factor influencing the region-and-annum based WFgreen.
At regional and growth period level, WFgreen correlates negatively with corresponding Y/A to different degrees, depending on geographical scales selected.
Within PingLin alone, at regional and growth period level, the WFgreen correlates negatively and moderately with corresponding Y/A, with the correlation coefficient being -0.538. However, Within MingJian alone, at regional and growth period level, the WFgreen correlates negatively and weakly with the corresponding Y/A, with the correlation coefficient being -0.141. Therefore, further studies are required to investigate the factors resulting in weak negative correlation between the region-and-growth-period based WFgreen and corresponding Y/A within MingJian.
In conclusion, at regional and growth period level, within PingLin alone, the moderately negative correlation between the WFgreen and the corresponding Y/A suggests that Y/A possibly plays a part in determining the region-and-growth-period based WFgreen.
At farm and annual level, the WFgreen correlates negatively to the corresponding Y/A to different degrees, depending on the selected geographical
60
scales. Within PingLin along with MingJian, at individual farm and annual level, WFgreen correlates negatively and moderately with the corresponding Y/A, with the correlation coefficient being -0.596. Within PingLin alone, at individual farm and annual level, WFgreen correlates negatively and moderately with the corresponding Y/A, with the correlation coefficient being -0.695. Within MingJian alone, at individual farm and annual level, WFgreen correlates negatively and strongly with the corresponding Y/A, with the correlation coefficient being -0.816. Therefore, at farm and annual level, within PingLin along with MingJian, or within either PingLin or MingJian, the moderate to strong correlations between the WFgreen and the corresponding Y/A suggest that the farm-and-annum based WFgreen is possibly affected by Y/A.
In conclusion, the presumption that Y/A is a possible factor influencing WFgreen is reasonable based on the moderate to strong correlations between WFgreen and Y/A at various temporal and geographical level, despite the weak correlation between WFgreen and Y/A at growth period level within MingJian.
Table 5.8.1 Yield per unit area (Y/A) by farm within PingLin (kg/ ) Farm title
Growth period Farm A Farm B Farm C Farm D Farm E Overall Spring Growth Period 0.043 0.062 0.021 0.053 0.013 0.036 Summer Growth Period 0.002 0.009 0.000 0.000 0.002 0.002 Fall Growth Period 0.000 0.015 0.000 0.000 0.003 0.002 Winter Growth Period 0.019 0.028 0.009 0.024 0.006 0.016 Throughout 2012 0.064 0.114 0.030 0.077 0.025 0.057
Table 5.8.2 Yield per unit area (Y/A) by farm within MingJian (kg/ ) Farm title
Growth period Farm F Farm G Farm H Farm I Farm J Overall Spring Growth Period 0.026 0.056 0.012 0.041 0.016 0.029 Summer Growth Period 0.068 0.146 0.03 0.000 0.02 0.068 Fall Growth Period 0.026 0.056 0.012 0.041 0.016 0.029 Winter Growth Period 0.018 0.039 0.008 0.029 0.011 0.021 Throughout 2012 0.139 0.298 0.062 0.112 0.063 0.147
5.9 Factors influencing annual yield per unit area (Y/A)
Annual yield per unit area of tea farms (Y/A) is mainly affected by cultivation
61
methods, harvesting methods, and tea tree species according to empirical experiences of tea farmers of PingLin and MingJian.
The empirical experience of tea farmers that the Y/A ratio of manual to mechanical harvesting is 30% is partially supported by the information of Farm B and Farm D for the following reasons. First, Ching- Hsin-Oolong was grown on both Farm B and Farm D. Second, conventional cultivation was completely
The empirical experience of tea farmers that the Y/A ratio of manual to mechanical harvesting is 30% is partially supported by the information of Farm B and Farm D for the following reasons. First, Ching- Hsin-Oolong was grown on both Farm B and Farm D. Second, conventional cultivation was completely