full papers proceeding The 9th International Conference on Environmental Engineering, Science and Management_Final

- 385 -

Table 5 Salivary Cortisol level Abnormal Cortisol range
Area n (%)

Mae Moh 17 (32.7)
Ban Don
Sop Pad 14 (63.6)

Total 20 (64.5)

51 (48.6)

Table 5 shows that Sop Pad was the highest abnormal cortisol range 64.5%., Ban Don 63.6% and Mae Moh
32.7%. The area that nearest power plant shown the highest prevalence. An adverse psychological
environment has been shown to influence basal cortisol levels and cortisol response to stress. This depends
on the adverse stimuli, but also varies across individuals and may be influenced by genetic (9).

Table 6 Correlation between depression parameters and study area

Salivary cortisol concentration Study Area P-value
PHQ 9 score Correlation Coefficient* 0.00**
PSQI score 0.105
0.396 0.014
-0.159
-0.238

*Spearman’s correlation
**Significant correlation p<0.05

The results showed that there was a significant positive correlation between study area and salivary cortisol.
The prevalence of depression is significantly higher in residents of rural areas (Table 6). People in rural areas were
more likely to have characteristics that strongly associated with depression, including poor health status and chronic
disease (1).

CONCLUSION
In three study areas including Ban Don, Sop Pad and Mae Moh subdistrict had shown the different

prevalence of depression and sleep quality. The distance between resident and power plant did not show

significantly difference. But three study area showed the correlation between Salivary cortisol concentration
among elderly and the nearest area of power plant, where “Sop Pad” was the highest number of abnormal

salivary range.

ACKNOWLEDGEMENT
The research reported here were supported in part of thesis at Collage of public health, Chulalongkorn
university, Bangkok, Thailand.

9th International Conference on Environmental Engineering, Science and Management
The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 386 -

REFERENCE
[1] Ancelin, M., Scali, J., Norton, J., Ritchie, K. Dupuy, A., Chaudieu, I., and Ryan, J. The effect of an

adverse psychological environment on salivary cortisol levels in the elderly differ by 5-HTTLPR
genotype. Neurobiology of stress. 2017; 7: 38-46.
[2] Buysse, D., Reynolds, C., Monk, T., Berman, S., and Kupfer ,D. The Pittsburgh sleep quality index:
A new instrument for psychiatric practice and research. Psychiatry Research. 1989;28(2):193-213.
[3] Halperin, D., Environmental noise and sleep disturbances: A threat to health. Sleep Science. 2014;7:
209-212.
[4] Kroenke, K., Spitzer, R., and Williams ,J. The PHQ-9 Validity of a Brief Depression Severlty
Measure. Journal of General Internal Medicine. 2001(16): 606-613.
[5] Laudat, M., Cerdas, S., Fournier, C., Guiban, D., Guilhaume, B., Luton, J. Salivary cortisol
measurement: A practical approach to assess pituitary-adrenal function. The journal of clinical
endocrinology & metabolism. 1988;66(2):343-8.
[6] Luppa, M., Sikorski, C., Luck, T., Weyerer, S., Villringer, A., König H., and et al. Prevalence and risk
factors of depressive symptoms in latest life--results of the Leipzig Longitudinal Study of the Aged
International Journal of Geriatric Psychiatry. 2012;27(13):286-295.
[7] Nations, U. World Population Ageing 2017 Highlights. In: Affairs DoEaS, editor. New York, 2017.
[8] Organization WHO. Mental health of older adults 2017 [Available from: https://www.who.int/news-
room/fact-sheets/detail/mental-health-of-older-adults.
[9] Probst, J., Laditka, S., Moore, C., Harun, N., Powell, P., and Baxley, E. Rural-Urban differences in
depression prevalence: Implication for family medicine. Health service research. 2004; 38(9);
653-660.
[10] Sidik ,S. , Rampal ,L., Afifi, M. Physical and mental health problems of the elderly in a rural
community of Sepang, Selangor. Malaysian Joutnal of Medical Sciences. 2004;11(1):52-59.
[11] Sozeri, G. Depression in the Elderly: Clinical Features and Risk Factors. Aging and Depression.
2012(3): 465-471.
[12] Thichumoa, W., Howteerakul, N., Suwannapong, N., and Tantrakul, V. Sleep quality and associated
factors among the elderly living in rural Chiang Rai, northern Thailand. Epidemiology and Health.
2018; 40: 1-9.

9th International Conference on Environmental Engineering, Science and Management
The Heritage Chiang Rai, Thailand, May 27-29, 2020

HIROSAKI

- 387 - H-01

Green Chromatic Coordinate (GCC) Affects CO2 Budget
in Beech Forest

Sachinobu Ishida1* Motomu Toda2 Tsukasa Saito1 Yuushi Igari1 and Daiyu Ito3

1*Graduate School of Science and Technology, Hirosaki University, Aomori 036-8561, Japan
2Graduate School of Integrated Sciences for Life, Hiroshima University, Higashi-Hiroshima 739-8521, Japan

3Faculty of Agriculture and Life Science, Hirosaki University, Aomori 038-3802, Japan
*Phone : +81-172-39-3621, E-mail : [email protected]

ABSTRACT
Net ecosystem production (NEP) of forest consists of gross primary production (GPP) by photosynthetic
carbon uptake, and ecosystem respiration (RE). Each component is strongly affected by both ecosystem and
meteorological conditions, moreover climate change and should be monitored continuously. Green chromatic
coordinate (GCC) is one of phenology index which is calculated using images taken by inexpensive interval
cameras. The objectives of this study are to evaluate GCC as an index of forest phenology and to clarify
relationships between GCC and CO2 budget observed in the Shirakami-Sanchi beech forest, heavy snowfall
mountainous area. As a result, seasonal variation of GCC detected the timing of snow disappearance, beech
phenology, and correlated well with photosynthetic CO2 uptake. Thus, GCC is one of the important indexes
to evaluate the CO2 budget of the vegetated area.

Keywords: CO2 budget; beech forest; green chromatic coordinate; phenology; snow; photosynthesis

INTRODUCTION
Forest ecosystems are important carbon dioxide (CO2) sink. Net ecosystem production (NEP) of forest
consists of gross primary production (GPP) by photosynthetic carbon uptake, and ecosystem respiration
(RE). Each component is strongly affected by both ecosystem and meteorological conditions, moreover
climate change (IPCC, 2013) [1]. GPP is controlled mainly by leaf area, photosynthetically active radiation
(solar radiation), and air temperature. RE is controlled mainly by biomass, soil temperature, soil water
content, and air temperature. The monitoring of both conditions is fundamental to understand the interaction
between them. NDVI, normalized difference vegetation index, using reflectance of photosynthetically active
radiation and solar radiation is one of the well-known indexes of detecting vegetation phenology. Another
index of phenology is green chromatic coordinate (GCC) which is calculated using images taken by
inexpensive interval cameras (Sonnentag et al., 2012) [2].

The objectives of this study are to evaluate GCC as an index of forest phenology and to clarify
relationships between GCC and CO2 budget in the beech forest. Distribution of beech forests in Japan
corresponds to heavy snow area. Snow amount and snow cover period are directly affected by climate
change, and affect forest phenology.

METHODOLOGY
Observation site was one of the AsiaFlux sites (SRK; http://asiaflux.net/index.php?page_id=106;
40°33´56"N, 140°7'40"E, 340 m a.s.l.) located in the Shirakami-Sanchi, 5 km away from World Natural
Heritage region, which is in Aomori prefecture, northern Honshu of Japan (Figure 1). The site is a cool-
temperate forest dominated by Japanese beech (F. crenata, 58% basal area) and Japanese big-leaf magnolia
(M. obovata, 19% basal area). The canopy height was 15-20 m and the maximum leaf area index (LAI) was
around 4.5. The site had annual mean precipitation 2,900 mm and a mean annual temperature 8.1℃, since
August 2008. The continuous snow cover period of the normal year was from December to April.

Meteorological and CO2 flux measurements were conducted at a 34m tower with solar panels and
batteries (Ishida et al., 2009) [3]. The tower stands on a 15 degree west faced slope. The fetch for the
prevailing wind direction of the site is approximately 1 km. Eddy covariance CO2 flux measurements were
conducted at 32 m height on the tower. Three orthogonal wind components and temperature were measured
using a 3D sonic anemothermometer (CSAT3; Campbell Sci.). The CO2 concentration was measured using
an open-path infrared CO2/H2O gas analyzer (LI-7500; LI-COR). All those data were recorded at 10 Hz in a

9th International Conference on Environmental Engineering, Science and Management
The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 388 -

data logger (CR3000, Campbell Sci.). Eddy covariance CO2 fluxe Fc was calculated every 30 minutes using
the EddyPro software (LI-COR) which included coordinate rotations and various corrections: Fc = <w’c’>,
where w is the vertical wind velocity, c is the CO2 density, prime means deviation from 30min average, and
< > means 30min average.

Figure 1 Location map of observation site

Gap filling of CO2 flux data was required to evaluate monthly values from 30-min data since missing
data existed. To fill the gaps, we calculated monthly (or semimonthly) relationships between RE and air
temperature Ta using nighttime flux data: RE = R0 exp[a (Ta – T0)], where R0 and T0 are mean RE and Ta
during snow cover period, and a is a least-square fitting coefficient. Relationships between photosynthetic

uptake speed (= GPP) and solar radiation S were also obtained using measured daytime Fc and estimated RE:
GPP = Fc – RE = [fS + Agmax – {(fS + Agmax)2 – 4fSqAgmax}0.5] / 2q, where f is the initial slope, Agmax is the
maximum gross photosynthetic speed, q is the convexity of the curve. In this least-squared fitting process, f,
q, and Agmax were obtained.

Canopy pictures were taken at 9:00, 12:00, 15:00 and 18:00 LT every day using an interval camera
(KADEC21-EYEⅡ, North one), which was mounted at 32 m height on the tower. Canopy greenness was

quantified using the green chromatic coordinate (GCC), which uses digital numbers of red (R), green (G),
and blue (B) to calculate the ratio of green within the interest image region: GCC = G / (R + G + B). We

calculated the GCC of every picture using PhenoCam ver1.1 (Sonnentag et al., 2012) [2], then averaged
every 3 days and month to evaluate the relationship with CO2 fluxes. The evaluation period is from May to
August during 2013-2016.

RESULTS AND DISCUSSIONS
Table 1 shows meteorological conditions of the evaluation period, May to August, during 2013-2016. In 2013,
continuous snow disappeared in the middle of May which was about 2 weeks later than other years. In 2014, the
site has much precipitation throughout the season and much solar radiation in May.

Table 1 Meteorological conditions of evaluation period, May to August

Year Air temp. Solar rad. Precip. Snow
Mean Mean Total Disappeared
2013 (℃) (W m-2) (mm)
2014 17.4 199 800.5 20 May
2015 1242.5 05 May
2016 17.8 235 628.5 04 May
17.8 684.0 21 Apr.
215
17.9
207

9th International Conference on Environmental Engineering, Science and Management
The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 389 -

Figure 2 Seasonal variation of GCC

Figure 3 Relationships between GCC and Agmax (upper panel), GPP (lower panel)
9th International Conference on Environmental Engineering, Science and Management

The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 390 -

Figure 2 shows 3-days averaged GCC variations of each year. GCC variations reflected beech phenology
well. New leaves of beech trees in this site opened just before snow disappearance and leaves were colored and
falling in October. The delay of spring bud burst in 2013 was also captured well by GCC (Figure 2). This delay
was caused by the slow snow disappearance of 2013.

Maximum monthly GPP and RE were recorded in July or August except for 2015 (Figure not shown).
Maximum monthly net ecosystem production (NEP = GPP − RE) and GPP were recorded in the same month
since the seasonal GPP variation range was larger than that of RE. Accordingly, understanding the variation of
GPP is essential for the evaluation of the CO2 budget of this forest. GPP is controlled by photosynthetic capacity
(i.e. light response curve of photosynthesis) and solar radiation. Figure 3 shows relationships between GCC and
GPP or Agmax. Both of GPP and Agmax were strongly correlated with GCC in each month and throughout the
season. Especially in May of 2013, small GCC value reflected small LAI.
CONCLUSION
Not only meteorological conditions but also vegetation phenology strongly affect the CO2 budget of the
vegetated area. We conduct meteorological and CO2 flux measurements, and canopy monitoring in
Shirakami-Sanchi beech forest, heavy snowfall mountainous area. As a result, seasonal variation of GCC
detected the timing of snow disappearance, beech phenology, and correlated well with photosynthetic CO2
uptake. Thus, GCC is one of the important indexes to evaluate the CO2 budget of the vegetated area.
ACKNOWLEDGEMENT
This work was supported by JSPS KAKENHI Grant Number JP25450201.
REFERENCE
[1] IPCC. 2013. Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to

the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge
University Press, Cambridge.
[2] Sonnentag, O., Hufkens, K., Teshera-Sterne, C., Young, A.-M., Friedl, M., Braswell B.-H., Milliman,
T., O’Keefe, J., Richardson, A.-D. 2012. Digital repeat photography for phenological research in
forest ecosystems. Agricultural and Forest Meteorology. 152: 159-177.
[3] Ishida, S., Ito, D., Matsuura, Y. 2009. Overview of Shirakami flux tower and general meteorological
conditions in July to October 2008. Shirakami Kenkyu. 6: 18-25. (in Japanese with English abstract)

9th International Conference on Environmental Engineering, Science and Management
The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 391 - H-02

The Effects of Aquaculture Wastes Application on Heavy Metal
Distribution in Apple Orchard Soils

Eriko Komori1 Chihiro Kato2* Hidekazu Kobatake3 Akira Endo4
Choichi Sasaki5 and Nobuhiko Matsuyama4

1*Graduate student, Graduate school of Agriculture and Life Science, Hirosaki University;
2Assistant Professor, Faculty of Agriculture and Life Science, Hirosaki University, Aomori, 036-8561,Japan

3 Associate Professor, Institute of Regional Innovation, Hirosaki University, Aomori, 030-0813, Japan
4Associate Professor, Faculty of Agriculture and Life Science, Hirosaki University
5 Professor, Faculty of Agriculture and Life Science, Hirosaki University

*Phone :+81-172-39-3796, Fax : +81-172-39-3796., E-mail : [email protected].

ABSTRACT
Aomori prefecture, Japan, is one of the large scallop producers in Japan. In scallop culture, plenty of small
shells and/or marine algae cling to the rearing cages and they are recognized as aquaculture wastes. Because
of its large quantity, those wastes often piled up out in the open for a while and cause odor problems.
„Hiatella orientalis“ has been recognized as a kind of the aquaculture wastes, which contains cadmium in
their inner organs. In this study, we investigated the effects of “Hiatella orientalis” application, as a soil
amendment, on soil pH and heavy metal distribution of Cu accumulated apple orchard soils by solute
transport experiments using soil columns. Individuals of “Hiatella orientalis” which were separated from the
collected aquaculture wastes were mainly consisted of their shells though they were not washed out. Then
dried “Hiatella orientalis” was ground with a mill into powdered form and was mixed with surface soils of
the columns. Then the irrigation as rainfall was conducted every three days for two months. Application of
powdered shell resulted in increase of soil pH, but only at the soil surface layer, where the powdered shell
were mixed. Soil EC increased by powdered shell application for the entire soil layer due to the soluble
chemicals originated from seawater. In this study, 0.1M HCl extracted Cd concentration was low or Cd was
almost not detected. Regarding to Cu, it was concentrated on the soil surface and powdered shell might help
prevent Cu be soluble by increasing soil pH.

Keywords : aquaculture waste; Hiatella orientalis; apple orchard; copper accumulation; soil amendment

INTRODUCTION
Aomori prefecture, located in the northernmost part of Honshu Island, is the second largest scallop

producer in Japan. In scallop culture, it is often found that plenty of small shells and/or marine algae cling to
the rearing cages, which are recognized as aquaculture wastes. Because of its large quantity, it takes time and
costs a lot for proper disposal. Therefore those wastes often piled up in the open for a while and cause odor
problems. „Hiatella orientalis“ has been recognized as a kind of the aquaculture wastes. Since it is reported
that their inner organs contain cadmium (Cd), the disposal and/or usage methods of these wastes have been
under discussion.

Aomori prefecture is also known as the largest apple producer in Japan. In apple orchards, Bordeaux
mixture, a mixture of copper sulfide and calcium carbonate has often been used as a pesticide. Therefore, it
has been found that, copper (Cu) concentrations of the soil surface layer in apple orchards are sometimes
higher than Japanese safety standard, 125 mgCu kg-1 (Inoue, 2007 [1]). Aoyama and Nagumo (1998) [2]
mentioned that in apple orchard soils with heavy metal accumulation, the carbon mineralization of humified
plant residue was inhibited by the pH-dependent water soluble and exchangeable heavy metal, especially Cu.
Since the undergrowth plays an important role in the cycling of nutrients in apple orchards, Cu accumulation
of apple orchard soils may have negative impacts on nutrient cycling through the undergrowth (Aoyama and
Nagumo, 1998, [2]). Since the solubility, or the availability, of Cu is high under low pH condition, alkali has
often been applied to heavy metal contaminates soil such as Cd and Cu (Guo et al., 2006 [3]). Liming
materials including CaCO3 are examples of the fixing additives. Seashells are typically composed of CaCO3
and are sometimes (especially Mizuhopecten yessoensis (Japanese scallop) and/or Crassostrea gigas
(Japanese oyster) shells) applied as lime to agricultural lands.

9th International Conference on Environmental Engineering, Science and Management
The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 392 -

From the above, the objective of this study is to clarify the effects of Hiatella orientalis, mainly their
shells, application, as soil amendments or alkali, on the distribution of heavy metals in apple orchards soil.

METHODOLOGY
The soil was collected from the upper 10 cm soil layer at the Apple Research Center of Aomori

Prefectural Industrial Technology Research Center, Aomori, Japan. Kuroboku Andisol was distributed in this

area, and the physical and chemical properties of the soil were shown in Table 1. Cation exchange capacity
(CEC) was measured by Schollenberger’s method (Schollenberger and Simon, 1945 [4]). The collected soil

was sieved with a 2.0 mm nylon sieve and then soil moisture was added to adjust the gravimetric water
content ω = 0.48 g g-1.

The aquaculture wastes were sampled from Mutsu bay, Aomori, Japan and only the individuals of
“Hiatella orientalis”, mainly their shells, were separated. The shells were not washed and were oven-dried
(105~110℃), and then they were ground by using a mill (WONDER CRUSHER WC-3) into powder form.

In this study, the solute transport experiment was conducted by using soil columns (polyvinyl chloride;
Diameter = 10cm, Height = 24cm) in the laboratory. Any plants were not grown in this experiment. The
bottom face of the column was covered with acryl plate with small holes and thus it allowed water to drip out

from the column when the bottom was under water saturated condition. From the bottom, 3cm-thick gravels
(φ2 ~ 4.75 mm; bulk density ρd = 1.5 g cm-3) and 18cm-thick sampled soil (ρd = 0.65 g cm-3) were packed.

The soil moisture samplers (Daiki Rika Kogyo, Co. Ltd.) were inserted at depths of 5, 11, 17 cm from the
soil surface in order to sample soil water and measure the sampled soil water pH. The soil moisture sensors,

EC5 (METER), were installed at depths of 8 and 14 cm from the soil surface.
Three models (two repeats for each) were prepared in the columns; (a) Control, (b) mixing the ground

shell powder in the surface soil layer (0 ~ 3 cm depth from the soil surface) (hereafter „Shell“, (c) adding
CuCl2 solution (1025 mgCu kg-1) and mixing the ground shell powder in the surface soil layer (Hereafter
„Shell +Cu“ ) (Figure. 1). The treatment (c) “Shell + Cu” were designed after Cu accumulated soil in apple
orchards (Inoue et al., 2007 [1]). The shells were applied to increase the soil pH into 6.0, the optimum pH for
apple growing, and the rate of the application (1.28 kg magnesia lime per square meter) was determined by

the guide of apple producing (Aomori Apple Association, 2015 [5]).
The irrigation, 8.7 mm day-1, was done every three days for two months as rainfall. Totally, 157.2 mm,

which was calculated as approximately 1.8 pore volume, of water was irrigated. The precipitation intensity
and frequency was determined by 5-year AMeDAS weather station data (Japan Meteorological Agency) of
non-snowy period at Hirosaki City, Aomori, Japan. The pH of irrigation water (simulated rain water) was

adjusted to pH 5.0 by using 1M nitric acid and 1M sulfuric acid because of the acidity of the rain in Aomori
prefecture (Kitamura et al., 1991 [6]).

Soil water sampling from each columns was conducted every nine days, the next day of the irrigation
and then measured soil water pH. The drained water from the bottom of the soil column was stored in a
plastic bottle and the bottle was changed to the new one every nine days when soil water sampling was

conducted.
After two months irrigation, each soil column was cut into 7 layers (3cm thickness for each layer)

including the gravel layer. Then the soil was air dried and soil pH (1:2.5 water extraction) and soil electrical
conductivity (EC; 1:5 water extraction) were measured. The vertical distribution of 0.1M HCl extracted
cations, Cd, Cu, Mg, Ca, and Na, of the soils were also obtained with atomic absorption spectrophotometer

(HITACHI Z-200).

Table 1 Physical and chemical properties of the soil

Depth Soil Dry bulk pH(H2O) Exchangeable cation CEC
0 ~ 10 cm texture (cmolc kg-1) (cmolc kg-1)
density
Loamy (g cm-3) Ca2+ Mg2+ K+ Na+
sand
0.70 4.76 1.12 0.63 0.10 0.59 21.0

9th International Conference on Environmental Engineering, Science and Management
The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 393 -

Soil 3 cm Soil + shell 3 cm Soil + Cu + shell
surface

18cm Soil Soil Soil

3 cm Gravel

(a) Control (b) Shell (c) Shell +Cu

Figure 1The experimental design of the three models

RESULTS AND DISCUSSIONS

Figures 2 and 3 show the vertical distribution of soil pH and 0.1 M HCl extracted Ca of soil,
respectively. While the soil pH at the surface layer (0 ~ 3 cm depth from the soil surface) increased to
approximately 6.0 by mixing the shell powder, soil pH values of lower layers were almost the same among
the three models (Figure 2). As well, though the water was irrigated about 1.8 pore volume, increase of soil
Ca concentration with shell application mainly occurred in the surface layer (0 ~ 3cm) and soil Ca
concentration in lower layer was almost the same regardless of the shell application. These results should
reflect the low solubility of CaCO3 of the shell.

Figures 4 (a) ~ (c) show the changes in soil water pH of each sampling depths of (a) “Control”, (b)
“Shell” and (c) “Shell + Cu” respectively, and Figure 4 (d) shows the changes in drained water pH of each
treatment. Both soil water pH and drained water pH were around pH 5.0 as irrigated water and it also
supports the poor solubility of the shell.

Soil pH Soil Ca concentration (mg kg-1)

4567 0 1200 2400 3600
0
0

33

Depth from the surface (cm)
Depth from the surface (cm)
66

99

12 Control 12 Control
Shell Shell

15 Shell + Cu 15 Shell + Cu
Initial soil pH
18
18
Figure 3 The vertical distribution of soil Ca
Figure 2 The vertical distribution of soil pH at concentration at the end of the experiment
the end of the experiment

9th International Conference on Environmental Engineering, Science and Management
The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 394 -

Figure 4 Changes in sampled soil water pH at the depths of 5, 11, and 17 cm; (a) control
(b) Shell, (c) Shell + Cu; and changes in drained water from the bottom of the soil column

Figure 5 shows the comparison of soil electrical conductivity (EC; 1:5 water extracted) between
“Control” and “Shell”. The soil EC increased with shell application for almost all the depths of the column.
It should be caused by soluble chemicals such as sodium (Fig.5 (b)) and magnesium salt originated from sea
water. MAFF (2008) [7] recommends the soil EC to be under 0.3 mS cm-1 for Kuroboku Andisol in upland
fields. The soil EC of the “Shell” in this study was 0.3 ~ 0.4 mS cm-1, so in case of field application, leaching
or excess water irrigation may be effective to maintain the proper EC values.

Figure 5 The vertical distribution of (a) soil EC and (b) Soil Na
concentration at the end of the experiment

9th International Conference on Environmental Engineering, Science and Management
The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 395 -

Figures 6 (a) and (b) show the vertical distribution of Cd and Cu, respectively. In this study, Cd was
detected only at depths of 1.5 and 4.5 cm from the soil surface of „Control“ and at a depth of 10.5 cm from
the soil surface of „Shell“. Furthermore, the soil Cd concentrations were lower than the average soil Cd
concentration of non-contaminated agricultural fields in Japan (0.33 mg Cd kg-1) (Asami, 2010 [8]) (Fig.6
(a)). Copper concentration at the surface layers was lower in „shell“ than in „Control“ and it is probably due
to the increase of soil pH. In the „Shell + Cu“ model, Cu concentration at the soil surface was approximately

1/10 compared with total added Cu and only little was infiltrated into the lower layer. This might be because

of the adsorption with clay and organic matter as well as the decrease of solubility accompanied with

increasing soil pH.

Soil Cd concentration (mg kg-1) Soil Cu concentration (mg kg-1)
0 0.1 0.2 0.3 0 50 100 150
0 0

(a) (b)

3 3

Depth from the surface (cm)
Depth from the surface (cm)
66

99

12 Control 12 Control
Shell
Shell Shell + Cu

15 Shell + Cu 15

18 18

Figure 6 The vertical distribution of soil Cu and soil Cd

CONCLUSION
In this study, we investigated the effects of “Hiatella orientalis”, a kind of aquaculture wastes, application on
soil pH and heavy metal distribution of Cu accumulated apple orchard soil. Individuals of “Hiatella
orientalis” which were separated from the collected aquaculture wastes were mainly consisted of their shells
though they were not washed out. Application of powdered shell resulted in increase of soil pH after two
months irrigation, but only at the soil surface layer, where the powdered shell was mixed. Soil EC increased
by powdered shell application for the entire soil layer due to the ions originated from seawater. In this study,
0.1M HCl extracted Cd concentration was low or Cd was almost not detected. Regarding to Cu, it was
concentrated on the soil surface and powdered shell might help prevent Cu be soluble by increasing soil pH.

ACKNOWLEDGEMENT
We appreciate Sotogahama Fisheries cooperative for their help in collecting aquaculture wastes. This study
is partly supported by Chokei Co. Ltd. (Aomori, Japan).

9th International Conference on Environmental Engineering, Science and Management
The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 396 -
REFERENCE
[1] Inoue, H., K. Masuda ， K. Sakamoto M. Nukada ， Y. Umemiya and M. Kita. 2007. Copper

ccumulation and Chemical Forms in Bordeaux Mixture Spraying Apple Orchard Soils, Jpn. Soil
Science and Plant Nutrition, 81-83
[2] Aoyama, M. 1998. Effects of heavy metal accumulation in apple orchard soils on the mineralization of
humified plant residues, soil science and plant nutrition, 44(2), 209 – 215
[3] Guo, G., Q. Zhou, L. Q. Ma. 2006. Availability and assessment of fixing additives for the in situ
remediation of heavy metal contaminated soils: A review, Environmental Monitoring and Assessment,
116, 513-528
[4] Schollenberger, C.J. and Simon, R.H. 1945. Determination of Exchange Capacity and Exchangeable
Bases in Soil-Ammonium Acetate Method. Soil Science, 59, 13-24.
[5] Aomori Apple Association, 2014, A guide to apple production 2014-2015, p133
[6] Kitamura, M. T. Kato, K. Sekiguchi, K. Taguchi, M. Tamaki, M. Oohara, J. Mori, K. Murano, S.
Wakamatsu, Y. Yamanaka, T. Okita, H. Hara. 1991. pH and Its Frequency Distribution Patterns of
Acid Precipitation in Japan, Journal of the Chemical Society of Japan, 6, 913-919
[7] MAFF, 2008 https://www.maff.go.jp/j/press/nousin/sekkei/pdf/110624-
01.pdf#search=%27%E9%99%A4%E5%A1%A9+%E8%BE%B2%E6%9E%97%E6%B0%B4%E7%
94%A3%E7%9C%81%27, Access: 1st April, 2020
[8] Asami, T. 2010. Toxic heavy metal pollution in soils of Japan. Agune-gijutsu center, Tokyo, Japan, p.7

9th International Conference on Environmental Engineering, Science and Management
The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 397 - H-03

Engineered Pseudomonas putida Strains Producing
cis,cis-Muconic Acid from Softwood Lignin-Related Phenols

Akihiro KIKUCHI1, Haruka SUGITA1, Minami HATAMURA1,
Miho AKUTSU1, and Tomonori SONOKI*

1Faculty of Agriculture and Life Science, Hirosaki University, 3 Bunkyo-cho, Hiroaski, Aomori 036-8561, JAPAN
*Phone : +81-172-39-3585, E-mail : [email protected]

ABSTRACT
The production of the materials that substitute for petrochemical-derived polymers from lignocellulose is one
of the technologies contributing to global warming. In this study, we aimed to create the bacterial strains
enabling the bioproduction of cis,cis-muconic acid (ccMA), a platform chemical for polymer industries, from
lignin-related phenols, and Pseudomonas putida KT2440-based two types of engineered strains were
developed. One was capable of producing theoretical yields of ccMA from the mixture of lignin-related
phenols although it still required other carbon sources such as glucose for growth and energy production. The
other strain was capable of producing ccMA without consuming any other carbon sources except lignin.

Keywords: cis,cis-muconic acid; lignin; Pseudomonas putida

INTRODUCTION
cis,cis-Muconic acid (ccMA) is a unique monomer
for use as a platform chemical in a variety of
polymer syntheses such as polyamide and
polyethylene via adipic acid and terephthalate
(Fig. 1), and also it is observed in aromatics
metabolic pathway of some bacteria. Thus, it has
been attempting to develop a biocatalyst capable of
producing ccMA efficiently from renewable
feedstocks such as sugars [1]. However, sugars are
in high demand for variety of bio-based products.
It is desirable to use lignin as a feedstock for the
production of value-added chemicals like
ccMA. It is generally described that lignin Fig. 1 ccMA is a base-molecule of various chemicals.
produces heterogeneous mixture of phenols by physicochemical and biological degradation and it is not easy
to obtain a specific compound in high yield. Whereas, microbial strains that can grow using these various
lignin-derived phenols retain a metabolic pathway that converts a mixture of these phenols by a specific
enzyme reaction system. For instance, Pseudomonas putida KT2440 is a metabolically versatile strain and
can use a variety of lignin-related aromatics as a carbon source for growth and energy production [2]. In the
metabolic pathway of KT2440, various aromatics from guaiacyl (G)-unit and p-hydroxyphenyl (H)-unit of
lignin are catabolized and funneled into protocatechuic acid (PCA) and further metabolized through the PCA
3,4-cleavage pathway [3]. Thus in this study, we aimed to develop P. putida KT2440-based engineered
strains that can produce ccMA from lignin-derived aromatic compounds efficiently.

METHODOLOGY
ccMA production from lignin-related phenols
PCA 3,4-dioxygenase (pcaHG) and ccMA cycloisomerase (catB) of P. putida KT2440 were disrupted (strain
IDPC, Fig. 2) [4]. The 0.2-kbp fragment containing the lactose promoter (Plac) from pUC118 was amplified
cloned into the NotI site of pJB866 [5] to generate pTS093, and then PCA decarboxylase (aroY, accession
no. AB479384), a flavin prenyltransferase (kpdB, AB920346)[6], vanillate O-demethylase (vanAB,
AE015041), p-hydroxybenzoate hydroxylase (pobA, AE015041), and catechol 1,2-dioxygenase (catA,
AE015041) were placed under the control of Plac sequence in pTS093 (designated as pTS094). IDPC was
transformed with pTS094 and the resulting strain, IDPC/pTS094, was precultured in 10 mL of F-medium
[6.03 g/L (NH4)2SO4, 3.88 g/L KH2PO4, 22.19 g/LNa2HPO4 12H2O, 2.90 g/L yeast extracts, 10 mL of metal

9th International Conference on Environmental Engineering, Science and Management
The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 398 -

solution. The metal solution was composed of 10.00 g/L MgSO4 7H2O, 500 mg/L FeSO4 7H2O, 75 mg/L
MgO, 200 mg/L CaCO3, 144 mg/L ZnSO4 7H2O, 112 mg/L MnSO4 5H2O, 25 mg/L CuSO4 5H2O, 28 mg/L
CoSO4 7H2O, 6 mg/L H3BO4, and 5.13 mL/L concentrated HCl] containing 15 g/L glucose, 25 mg/L
nalidixic acid (Nal), 50 mg/L gentamycin (Gm), 25 mg/L kanamycin (Km) and 20 mg/L tetracycline (Tc) at
30oC for 16 h, and 10 mL of preculture was used to inoculated 1 L of F-medium containing 15 g/L glucose
and the antibiotics in a 2 L vessel of mini-jar fermentor system (BMJ-02, ABLE Corporation) and incubated
at 30oC, pH 7.0 with agitation at 650 rpm and aeration at 1.0 L/min. When ccMA-producing ability of the
strain was evaluated in batch mode, ferulic acid (FA), p-coumaric acid (CA), vanillic acid (VA) or p-
hydroxbenzoic acid (HBA) was fed to the culture after 8 h from the inoculation. For fed-batch mode,
glucose, (NH4)2SO4, and lignin-related phenols (FA, CA, VA, HBA, or the mixture of VA and HBA) were
fed according to the consumption rates after 8 h from the inoculation. All aromatics were purchased from
Fujifilm Wako Pure Chemical Corporation. The concentrations of ccMA and other metabolites were
measured using a high-performance liquid chromatography (HPLC) system (1200 series, Agilent
Technologies Inc.) equipped with a ZORBAX Eclipse Plus C18 column (5 μm particles, 4.6 mm in diameter,
150 mm in length, Agilent Technologies) run at 40C with a mobile phase gradient [Solvent A: 5% (v/v)
CH3OH and 1% (v/v) CH3COOH in H2O; Solvent B: 50% (v/v) CH3OH and 1% (v/v) CH3COOH in H2O;
the latter was introduced after the injection of samples and ramped from 0 to 20% in the first 8 min, ramped
up to 100% over the next 5 min, and maintained for 5 min]. The flow rate of the mobile phase was 1.0
mL/min, and the detection wavelength was 280 nm. Glucose concentration was measured using a Biosensor
(BF-5, Oji Scientific Instruments Co., Ltd.).

Glucose-free ccMA production from lignin-related phenols
ccMA production in a testing tube scale
AroY and pcaHG were ligated with pTS093 to construct pTS110 and IDPC was transformed with pTS110.
IDPC/pTS110 was incubated in shake-culture at 30oC for 16 h in 5 mL of Lysogeny Broth (LB) containing
the antibiotics. A 50 μL portion of the preculture was inoculated into 5 mL of MM containing the lignin-
related aromatics and the antibiotics, and shake-cultured at 30C. The lignin-related aromatics used as a sole
carbon source were VA, HBA, or their mixture. The optical density at 600 nm (OD600) was monitored with
miniphoto 615R (TAITEC Co., Saitama, Japan), and the concentrations of ccMA and PCA in addition to VA
and HBA in the culture were quantified using HPLC as described above.

Whole-cell reaction using an aroY- and pcaHG-expressing recombinant of IDPC
IDPC/pTS110 was inoculated to 10 mL of LB containing 25 mg/L Nal, 50 mg/L Gm, 25 mg/L Km, and 20
mg/L Tc, and shake-cultured at 30oC until the OD600 reached at 1.0. The cells were collected by centrifuge
and washed twice with saline, and then resuspended in 9 mL of Modified M9 (MM)[7]. The suspension was
dispensed into three 15 mL tubes, in 1, 2.5, 5 mL each, and VA, HBA, or PCA was added (10 mM), and then
incubated at 30oC. The consumption of the aromatics and production of ccMA were quantified using HPLC
as described above.

Effect of aeration on in vivo PcaHG and AroY activity
PcaHG and aroY were separately inserted into pTS093 to construct pTS108 and pTS109, and IDPC was
transformed with each plasmid (designated as IDPC/pTS108 and IDPC/pTS109, respectively). Each
transformant was precultured in 10 mL of LB containing antibiotics and shake-cultured at 30oC for 16 h. The
preculture was inoculated in 1 L of MM containing 15 g/L glucose and appropriate antibiotics in a 2 L vessel
and incubated at 30oC, 1 L air/min, pH7.0 under the controlled dissolved oxygen (DO) concentrations (5% or
20%). The DO sensor was calibrated with 5%(w/v) sodium sulfite solution (as concentration of 0%) and the
air-saturated MM (as 100%) before incubation. Eight hours after the inoculation, 1.5 g/L PCA was added and
monitored the conversion (consumption) of PCA to evaluate the effect of DO on the PCA conversion by
AroY and PcaHG.

Dissolved oxygen (DO)-stat fed-batch culture for ccMA production from lignin-related phenols
VanAB and catA in addition to pcaHG and aroY were introduced into IDPC to generate IDPC/pTS119. First,
the effects of DO concentrations from 2.5 to 20% on the yield and productivity of ccMA were evaluated in
batch mode. The strain was precultured as described above, and 10 mL of the preculture was used to
inoculate into 1 L of MM containing (4.2 g/L VA, 3.5 g/L HBA, and the antibiotics) and incubated at 30oC,

9th International Conference on Environmental Engineering, Science and Management
The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 399 -
pH 7.0, 1 L air/min, with agitation (varied from 100 to 650 rpm). The concentrations of VA, HBA, PCA,
ccMA, and glucose were measure as described above. Next, ccMA productivity of IDPC/pTS119 was
evaluated in a fed-batch mode as follows. The strain was precultured in 10 mL of LB containing the
antibiotics and sake-cultured at 30oC for 16 h. The preculture (10 mL) was used to inoculated in 1 L of MM
containing 4.2 g/L VA, 3.5 g/L HBA, and antibiotics and incubated (30oC, pH 7.0) under the controlled DO
concentration (2.5%) with varied agitation (100 to 650 rpm). When OD600 of the culture was reached at 0.8,
the mixture consisting of 168.1 g/L VA,138.1 g/L HBA, and 132.1 g/L (NH4)2SO4 was fed to the culture at
the rate according to the consumption of VA and HBA. The concentration of ccMA and other metabolites
were quantified as described above.

(a)

(b)

(c)

(d)

9th International Conference on Environmental Engineering, Science and Management
The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 400 -
(e)

(f)

Fig. 2 An overview of the engineered metabolic pathway of IDPC (a), IDPC/pTS094 (b), IDPC/pTS108
(c), IDPC/pTS109 (d), IDPC/pTS110 (e), and IDPC/pTS119 (f).
AroY, PCA decarboxylase; CatA, catechol 1,2-dioxygenase; CatB, ccMA cycloisomerase; KpdB, flavin
prenyltransferase; PcaHG, PCA 3,4-dioxygenase consisting of PcaH (PCA 3,4-dioxygenase beta subunit)
and PcaG (alpha subunit); PobA, HBA monooxygenase; VanAB, VA O-demethylase consisting of VanA
(VA O-demethylase oxygenase component) and VanB (oxidoreductase component). BA, benzoic acid; CA,
p-coumaric acid; CL, catechol; ccMA, cis,cis-muconic acid; FA, ferulic acid; HBA, 4-hydroxybenzoic acid;
PCA, proocatechuic acid; VA, vanillic acid.

RESULTS AND DISCUSSIONS
High yield ccMA production from lignin-related phenols.
As P. putida KT2440 is capable of degrading various aromatic compounds including FA, VA (G-lignin
models), CA, and HBA (H-lignin models) via PCA 3,4-ring cleavage reaction and also metabolizing catechol
via ccMA, the production of ccMA from such lignin-related aromatics is expected by the inactivation of the
degradation of both PCA and ccMA and the introduction of PCA decarboxylation. As expected, the

Table 1 ccMA yields (% of the theoretical yield) from each lignin-related aromatic compound

Amounts of the aromatics added to the culture

1 g 5 g 10 g 20 g

CA 93.2±2.3 100.1±3.0 105.8* 97.3*
FA 100.0±1.1 103.0±2.5 105.5* 112.7*

HBA 104.4±1.1 100.6±1.8 - 97.9±5.7

VA 102.1±2.4 97.9±2.5 - 105.3±3.0

The values are the means ± standard deviations from triplicates experiments. *The values are

from single experiment.

engineered strain, IDPC/pTS094, showed theoretical ccMA-producing ability from each lignin-related
aromatic compound (Table 1). Then further evaluation of the ccMA-producibility from the mixture of VA
and HBA in fed-batch mode was conducted. It was revealed that IDPC/pTS094 was able to produce ca. 35
g/L ccMA from the mixture of lignin-related phenols (Fig. 3). Some other ccMA productions from lignin-
related phenols have been reported whereas, in those reports, a single phenolic compound was used as

9th International Conference on Environmental Engineering, Science and Management
The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 401 -

feedstock for ccMA production and no mixtures of lignin-related phenols have been used. Whereas, recently,
several engineered strains have been examined the ccMA-producing ability from lignin streams containing
variety of lignin-derived phenolic compounds generated as a result of physicochemical depolymerization [8].
The ccMA-producing strain developed in this study will also be subjected to the same evaluation.

Fig. 3 fed-batch ccMA production from lignin-
related phenols.
IDPC/pTS094 was cultured in F-medium containing
glucose as a carbon source for growth and energy as
described in the Methodology section.

Glucose-free ccMA production from lignin-related phenols.
We constructed a ccMA-producing strain, IDPC/pTS094, that enables producing ccMA from lignin-related
phenols in high yield as shown above. This strain was deficient in the assimilation of not only ccMA but also
PCA (Fig. 2a) so other carbon sources such as glucose were required for growth and energy. Glucose and
other easily available carbon sources are already in high demand of usage in food production and thus it is
predictable to become competing to the obtained glucose for ccMA production. Then, we next aimed to
design the metabolic pathway that enables the bacterial strain to grow without consuming any other carbon
sources except lignin-related phenols and produce ccMA. The simultaneous expression system of pcaHG
and aroY IDPC was introduced into IDPC (IDPC/pTS110). As shown in Fig. 4, the strain grew on VA and/or
HBA

(a) (b)

(c)

Fig. 4. ccMA production from VA (a), HBA (b),
and the mixture of VA and HBA (c) by
IDPC/pTS110.
Each value is mean and standard deviation from
triplicates experiments. Symbols; closed circles,
ccMA; closed diamond, HBA; closed squares, VA;
closed triangles, PCA; open circle, OD600.

without any other carbon sources and produced ccMA in the yields of 2.9% (w/w, from VA, Fig. 4a), 12.5%
(w/w, from HBA, Fig. 4b), and 17.7% (w/w, from the mixture of VA and HBA, Fig. 4c), respectively. To
increase the yield of ccMA, we evaluated the effect of aeration on the productivity using the whole-cell
reaction under the different aeration conditions because PcaHG is an oxygenase and AroY has been reported
to be an oxygen-sensitive enzyme although the cofactor of AroY requires oxygen to get activated [9]. As
shown in Fig. 5, it was suggested that ccMA yield would be increased by reducing PcaHG activity under

9th International Conference on Environmental Engineering, Science and Management
The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 402 -
lower aeration condition. To evaluate which enzyme activity was affected in vivo by aeration, IDPC/pTS108
(expressing aroY) and IDPC/pTS109 (expressing pcaHG) were separately cultured under the DO-controlled
condition (Fig. 6). The consumption rate of PCA by IDPC/pTS108 (expressing pcaHG) in 5% DO-stat

(a) (b)

(c) (d)

Fig. 5 ccMA yields from VA (b), HBA (c), and PCA (d)
Portions (1.0, 2.5, and 5.0 mL) of the suspension of IDPC/pTS110 were dispensed into 15 mL tubes,
respectively, as depicted (a). The values are means from duplicates experiments.

(a) (b)

Fig. 6 Effects of DOs on in vivo activity of PcaHG and AroY
PcaHG expressing strain, IDPC/pTS108 (a), and aroY expressing strain, IDPC/pTS109 (b), were cultured in
5% (solid line) and 20% (dotted line) DO conditions. The results of IDPC/pTS109 in 20% DO condition are
means and standard deviations from triplicates experiments. Others are means from duplicates experiments.
Culture was lower than that observed in 20% DO (Fig. 6a). It is indicated that in vivo PcaHG activity is
sensitive to DO in PCA conversion. On the other hand, IDPC/pTS109 (expressing aroY) did not show clear
differences in PCA consumption rate as observed in IDPC/pTS108 but the variation became greater when it
was incubated in 20% DO condition (Fig. 6b). That means relatively low DO condition is suitable for in vivo
AroY activity. From these results, it was revealed that ccMA is producible from lignin-related phenols
without consuming any other carbon sources and DO-stat culture is effective to increase ccMA yield.

9th International Conference on Environmental Engineering, Science and Management
The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 403 -

Therefore, we cultured a ccMA-producing strain, IDPC/pTS119 in which vanAB and catA were expressed
along with pcaHG and aroY, in a batch mode and evaluated the effect of DO on ccMA production from the
mixture of VA and HBA (as a model of lignin-related phenols), and it was found that 5% DO was optimal to
yield higher ccMA amount stably (Fig. 7). Then, a DO-stat fed-batch culture of IDPC/pTS119 using the
mixture of VA and HBA as carbon sources for growth and energy was conducted. And it was confirmed that
10 g/L ccMA was produced without using any other carbon sources (e.g., glucose) except lignin (Fig. 8).

Fig. 7 Effect of DO on ccMA yields Fig. 8 Glucose-free ccMA production from
Aeration speed was maintained at 1 L air/min. lignin-related phenols by IDPC/pTS119.
Batch cultures were repeated 4 times in 2.5%
DO condition and, in the rest DO conditions,
repeated 2 times.

CONCLUSION
We achieved (i) development of a bacterial strain capable of producing ccMA in high yield from lignin-
related phenols and (ii) development of a ccMA-producing strain without consuming sugars. Further efforts
to develop a bacterial strain capable of funneling heterogeneous lignin-related phenols to value-added
chemicals would be significant. We are going to apply this concept to develop the microbial reactions that
produce not only aliphatic compounds but also aromatics for industrial use and lead the activity to produce
biomass-based and emerging polymers that will be able to substitute for petrochemical-derived polymers
from lignin.

ACKNOWLEDGEMENT
This work was supported by Advanced Low Carbon Technology Development program, Japan Agency for
Science and Technology (JST-ALCA, JPMJAL1506).

REFERENCE
[1] Myriant Corporation. Improved muconic acid production from genetically engineered

microorganisms. WO 2017/151811.
[2] Wackett, L.P. 2003. Pseudomonas putida--a versatile biocatalyst. Nature. 21(2): 136-138.
[3] Nelson, K.E., Weinel, C., Paulsen, I.T., Dodson, R.J., Hilbert, H., Martins dos Santos, V.A.P., Fouts,

D.E., Gill, S.R., Pop, M., and Holmes, M. 2002. Complete genome sequence and comparative analysis
of the metabolically versatile Pseudomonas putida KT2440. Environ. Microbiol. 4(12): 799-808.
[4] Sonoki, T., Takahashi, K., Sugita, H., Hatamura, M., Azuma, Y., Sato, T., Suzuki, S., Kamimura, N.,
and Masai, E. 2018. Glucose-free cis,cis-muconic acid production via new metabolic designs
corresponding to the heterogeneity of lignin. ACS Sustain. Chem. Eng. 6(1): 1256-1264.
[5] Blatny, J. M., Brautaset, T., Winther-Larsen, H.C., Karunakaran, P., and Valla, S. 1997. Improved
Broad-Host-Range RK2 vectors useful for high and low regulated gene expression levels in gram
negative bacteria. Plasmid. 38(1): 35−51.
[6] Sonoki, T., Morooka, M., Sakamoto, K., Otsuka, Y., Nakamura, M., Jellison, J., and Goodell, B. 2014.
Enhancement of protocatechuate decarboxylase activity for the effective production of muconate from
lignin-related aromatic compounds. J. Biotechnol. 192: 71-77.

9th International Conference on Environmental Engineering, Science and Management
The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 404 -
[7] Vardon, D.R., Franden, M.A., Johnson, C.W., Karp, E.M., Guarnieri, M.T., Linger, J.G., Salm, M.J.,

and Strathmann, T.J., and Beckham, G.T. 2015. Adipic acid production from lignin. Energy Environ.
Sci. 8(2): 617−628.
[8] Xu, Z., Lei, P., Zhai, R., Wen, Z., and Jin, M. 2019. Recent advances in lignin valorization with
bacterial cultures: microorganisms, metabolic pathways, and bio-products. Biotechnol. Biofuels.
12: 32.
[9] White, M.D., Payne, K.A.P., Fisher, K., Marshall, S.A., Parker, D., Rattray, N.J.W., Trivedi, D.K.,
Goodacre, R., Rigby, S.E.J., Scrutton, N.S., Hay, S., and Leys, D. 2015. UbiX is a flavin
prenyltransferase required for bacterial ubiquinone biosynthesis. Nature. 522(7557): 502-506.

9th International Conference on Environmental Engineering, Science and Management
The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 405 -

H-04

Environmental Impact of Riverbank Formed by Contaminated Soil

Sachi A. Wakasa1*, Vladan Marincović2, Tomomi Takeda3, Junichi Kurihara4, Lidja Đurđevac Ignjatović2,
Tamara Urosević2, and Renata Kovacević2

1* Assistant Professor, Institute of Regional Innovation, Hirosaki University, Aomori 030-0813, Japan.
2 Researcher, Mining and Metallurgy Institute Bor, Bor 19210, Serbia.
3 Researcher, Japan Space Systems, Tokyo 105-0011, Japan.

4 Associate Professor, Faculty of Science, Hokkaido University, Hokkaido 060-0810, Japan
*Phone : +81-17-762-7294, Fax : +81-17-762-5411, E-mail : [email protected]

ABSTRACT
This study investigated the forming of a polluted riverbank in a river with an upstream mine that has been
excavated for more than 100 years in Serbia. For this purpose, a field survey was conducted, and ground
surface sediments from rivers were collected, sediment thickness and density were measured, and soil pits
were excavated. Two soil pits were dug, their insides were observed, and sediment samples were collected.
The surface sediments collected were subjected to mineral identification using XRD, and the sediments
collected from the pits were subjected to chemical analysis. A high-accuracy orthoimage was created by
using unmanned aerial vehicles (UAVs) to acquire visible images with each overlapping 60% or more, and a
high-accuracy global positioning system (GPS) acquired position information. From this image, the area of
the riverbank was measured using geographic information system (GIS) software. Furthermore, to acquire
hyperspectral images, a camera was installed in the UAV. The obtained image analysis results showed that
pollutants from the mine were distributed on most of the surface of this riverbank. The study concludes that a
high possibility exists that the waste material from the mine accumulated on the riverbank, and the area is
96,000m2, the volume is 78,600 m3, and the weight is 152,400 t.

Keywords : contaminated soil; heavy metal; waste management; field survey; remote sensing;
jarosite; copper

INTRODUCTION
Contaminated soil is produced in mines and under geological conditions. Especially in the vicinity of mines,
the pollutants are suddenly discharged and diffused from the start of excavation, causing serious
environmental pollution. The forces of water and wind move the polluted soil, discharged by mining, on the
ground surface. Contaminated soil carried by river water behaves similarly to other debris [1]. However,
without knowing the status of the deposited pollutants, discussing the environmental pollution status, future
diffusion potential, environmental impact, and measures to improve them is impossible. Therefore, this study
reports the results of field surveys and environmental surveys using unmanned aerial vehicles (UAVs) for the
current state of river sediments that were transported and deposited by river water around the mine.

STUDY AREA
The city of Bor and its surroundings in eastern Serbia are among the largest mine in Europe and are well
known for copper deposits (Fig. 1). Around these deposits, since 1903, many current and historical copper
mines were developed. Our study area is the floodplain of a river located downstream of the Bor mine. The
Bor mine was mined upstream of the Bor and Krivelj Rivers and started in 1903 and 1979, respectively.
They are called the Old-Bor mine and the Veliki Krivelj mine (Fig. 1b). The study area is a riverbank formed
at the confluence of these rivers (Fig. 2a, c).
Underground mining started at the Old-Bor mine in 1903, and open-pit excavation began in 1923. Drilling
was continued until 1993. During that period, mining waste was directly discharged into rivers until 1933.
After that, overburden, low-grade ore and tailing were deposited around the mine. At the Veliki Krivelj
mine, excavation of the open pit started in 1979, and overburden, low-grade ore and tailing are deposited
around the open pit. The direct disposal to the river at the beginning of the excavation at the Old-Bor mine is
a problem. Conversely, although it can significantly affect the water quality environment, the subsequent
deposition of overburden and low-grade ore around the mine has not been included in the downstream waste
management in the material cycle management. However, regarding the tailings, the dam constructed as a

9th International Conference on Environmental Engineering, Science and Management
The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 406 -
sedimentation site builds up the tailings and is not fixed or covered (Fig. 2b). It’s easy to imagine what
behavior will be caused by rainfall or river flow.
In the study area, rivers with debris are deposited on the Bor and Krivelj Rivers in front of and after the
confluence, and a wider riverbank is formed (Figs. 2 and 5). Figure 2a shows the riverbank formed along the
Bor River, and Figure 2c shows the confluence of the Bor and Krivelj Rivers.

Figure 1 Study Area

Figure 2 The Study Area. (a) Bor River Riverbank (b) Tailings dam of Old-Bor Mine and (c)
confluence of the Bor and Krivelj Rivers

METHODOLOGY
To clarify the current condition of the target river basin, we conducted a field survey, analyzed the collected
samples in the laboratory, and analyzed the images acquired by UAV.
In the field survey, surface material was sampled, and the mineral composition of the samples was
investigated using X-ray diffraction (XRD) analysis. Riverbank sediments that developed along each river
were excavated, and sedimentary facies of the pits were observed. The sediment samples were also collected
from the excavated pits and analyzed for chemical content by ICP-MS and FAAS. The depths of sediments

9th International Conference on Environmental Engineering, Science and Management
The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 407 -

were investigated using a cone penetration test. By measuring the weight of the ground surface sample in a
fixed form on the spot, the density was obtained.
Two types of images were acquired using UAVs. One is a set of visible images in which adjacent places
overlap to create a high-precision digital topographic map. This set was taken using a GoPro camera
mounted on Phamtom3. The other set is hyperspectral images to classify the reflection properties of surface
materials. In total, 46 hyperspectral images were acquired every 10 nm in the range of 600–1050 nm. A
camera with liquid crystal tunable filter (LCTF) was used for the image acquisition [2]. The aerial
photography performed with the LCTF camera mounted on the UAV was conducted on 6 September 2018.
To create a high-precision topographic map, the acquired photographs are made into point clouds using
photo composition software (PhotoScan, Agisoft). The UAV also provided the position information but it
was corrected by the ground control point (GCP) acquired by a high-precision GPS locally (Trimble R4).
After correction, the area of the riverbank was measured from the created orthoimage using ArcGIS
software (ESRI).
The photographs taken by the LCTF camera were corrected geometrically for 46 images of each set and then
normalized using a gray mat corrected in the laboratory. Finally, the two sets were georeferenced and
merged. From the merged image, a spectral curve can be obtained for each pixel. However, correlating with
the spectrum curve obtained by the existing spectrometer was challenging. Conversely, a difference was
found in the spectrum for each pixel, and performing spatial analysis on the same image was possible.
Therefore, training data were acquired from the obtained image, and supervised classification were
performed by adopting the Spectral Angle Mapper (SAM) method.

RESULTS AND DISCUSSION
XRD of surface samples revealed that it contains jarosite, which is often found around mines where pyrite is
produced. Therefore, as shown in Figure 3, circle color shows the quantity of jarosite at the sampling point: large
content (L for red), medium content (M for green), and small content (S for blue). Jarosite is a secondary mineral
of iron oxide minerals and is a recrystallized product of mine waste dissolved in water [3]. Furthermore,
since the water used for crystallization is around pH2 [3], it indicates that acidic water is present in the
environment where jarosite is produced.

Figure 3. Field survey data on the orthoimages created in this study

Soil pits were dug on the left bank of the Bor River and the right bank of the Krivelj River. Figure 4 shows
each of the soil faces and the concentrations of copper, lead, arsenic, and manganese measured by ICP-MS
and iron and sulfur concentrations measured by FAAS for each sedimentary structure. Six sedimentary layers
are present in the pit of the Bor River, and alluvial deposits were observed beneath them. The concentration
of alluvium is lowest. The concentration of the second layer of the Bor Riverbank is missing and not shown.
In the sixth layer at a depth of 62–120 cm, sub-angular gravels with a maximum diameter of 20 cm were
observed, and the concentrations of copper, iron and sulfur were high at 1,900 ppm, 17%, and 27%,
respectively. Lead and arsenic exhibited the highest concentrations in the uppermost layer at 220 ppm and

9th International Conference on Environmental Engineering, Science and Management
The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 408 -

240 ppm, respectively. Manganese concentration was highest in the second layer. In the uppermost layer,
copper and iron were also generally high in concentration at 550 ppm and 4%, respectively.
As mentioned above, at the Old-Bor mine, for about 30 years after the start of mining in 1903, mining waste was
directly discharged to rivers. The 20 cm-sized gravel found at the sixth layer of the Bor Riverbank deposit is waste
at that time. This can also be explained by the extremely high concentration of copper, iron, and sulfur in the
bottom layer. The mining records of the Old-Bor mine indicate that the copper concentration was 6% for the first
30 years [4]. The presumption exists that gravel with such a large diameter flowed down since early 1933, as
waste was deposited on dams and sedimentation sites. Conversely, in the upper deposit, the grain size of sand to
silt is accumulated in layers, and the grade structure is seen in the fourth layer. These deposits could have flowed
down after the tailing dam was formed and when the dam wall collapsed because of a flood. The depth profiles of
the concentrations, except for manganese and sulfur, are similar. It can be inferred that this sediment does not
move much physically from the time of deposition and represents the situation at the time of deposition.
Manganese and sulfur are easier to move than iron. Especially in the first to third layers, the amount of sulfur is
relatively small, suggesting that the upper layer is being oxidized. Further, the fourth and fifth layers demonstrated
a high sulfur concentration, and their ratios to iron were 1.1 and 0.7. In other words, the area around this depth is
in an anaerobic state, indicating that the oxidation of pyrite has not progressed, indicating that the area around this
layer is the water table.
Six sedimentary layers were also observed in the Krivelj River pit, but the structure of the sediments was
significantly different from that of the Bor River. Furthermore, because of the water table, digging into a layer
below the bottom layer was impossible. The pit deposits on the Krivelj River were fine, mostly clay to silt size.
Strong elasticity and plasticity were observed in the dark gray clay layer at the bottom. Even in the Krivelj River
pit, the concentration of copper in the bottom layer was the highest at 1,400 ppm. Conversely, the concentrations
of iron and arsenic were high in the middle, white-gray silt layer at 8.9% and 230 ppm, respectively. Manganese
and sulfur concentrations did not differ from those found in Bor. The iron.sulfer (Fe/S) ratio was 1 or less than 1
in most layers. The Fe/S ratio was 2.9 only in the layer between 53–80 cm. The color of this layer is brown,
consistent with the oxidation and sulfur elution that would have occurred.
The sediments of the Krivelj Riverbank are even finer. This is convincing evidence that the tailings produced
during excavation and beneficiation after 1979, when technological innovation passed, might have flowed down.
The investigation of these particle sizes was only visual and could not be accurately measured. The difference in
the concentration of each layer could be interpreted by comparing it with the drilling record. However, the
behavior of heavy metals in the soil must also be considered, which was not done in this study. We want to make
it a future issue.
Both the Bor and Krivelj Riverbanks exhibit six sedimentary layers. To interpret this, a detailed topographical
survey must be conducted. Researchers estimated that the Bor Riverbank deposits were brown and exhibited less
sulfur than when they were deposited, indicating that enough time elapsed for oxidation. In the Krivelj Riverbank,
white and gray deposits are present, suggesting that they have not yet been oxidized. Bor deposits should be over
100 years old, and Krivelj deposits should be 40 years old. Several reasons exist for the difference in the
deposition period and the same number of layers. (1) During the first 70 years, deposits from the Old-Bor mine
were deposited on the Bor Riverbank. Consequently, the riverbed of the Krivelj River was lowered, and its
riverbank became prone to deposits. (2) Since the Krivelj River exhibits a larger catchment area than the Bor
River, sediments are more likely to collect. (3) The Veliki Krivelj mine underwent technological innovation and
large-scale construction. Therefore, the amount of sediment increased rapidly, and a large amount of sediment
came to flow suddenly. (4) Much bedding is visible in the sediments of the Bor Riverbenk, and other beddings
could exist that cannot be distinguished in this observation. Nevertheless, because all the layers contain chemical
substances considered to be of mining origin, it can be inferred that all of these deposits were mining waste.
As observed in the two soil pits, the deposit thickness was 120 cm and deeper than 100 cm. Furthermore, the
thickness of sediments in the riverbank was examined by a portable corn penetration tester. The survey
points are the purple hexagonal points shown in Figure 3, and the results are the numbers written at each
point. The depth near the river channel was deep, and the one far was shallow. The average of the Bor
riverbank was 70 cm, and the one of the Krivelj and after junction riverbank was 84 cm. Overall, the average
sediment depth was 76 cm. The density of surface sediments was also measured in the riverbank of the Bor
River, the Krivelj River, and after the junction with values of 1.76 g/cm3, 1.76 g/cm3, and 1.97 g/cm3,
respectively.

9th International Conference on Environmental Engineering, Science and Management
The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 409 -

9th International Conference on Environmental Engineering, Science and Management
The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 410 -

Figure 3 shows the orthoimage created from the visible images. This image was position-corrected with 15
GCP points measured by a precision GPS sensor. In this study, we estimated the area of riverbank sediment
from this orthoimage. Polygons were created with the area surrounded by the red line in the figure as
riverbank sediment, and the area of the polygon was calculated. Consequently, the total area of riverbank
sediments in this area was estimated at 96,000 m2.
The volume of riverbank sediment was calculated based on the sediment density, the average depth and the
area of the sediment from the field surveys in this study. The volume of the riverbank was calculated by
dividing it into the Bor Riverbank, the Krivelj Riverbank, and the riverbank after confluence. The result was
calculated as 18,400 t, 2,600 t and 131,400 t, respectively, in total, 152,400 t. The volume was 78,600 m3. In
other words, if contaminated soil for environmental protection must be covered, to cover the ground surface
would be 96,000 m2 or more. Also, if contaminated soil must be removed, 152,400 t of soil must be
transported and deposited.
Ninety-two (92) hyperspectral images were georeferenced and color corrected. Consequently, spectral data of 600
to 1,050 nm were obtained for each pixel. Figure 5 shows the obtained spectrum data from the images and the
ground truth samples. The ground truth samples indicated that the stronger the jarosite peak intensity during XRD,
the higher the amount of jarosite. The medium amount is jarosite M, and the small amount is jarosite S. The
spectral curves of the ground truth samples in Figure 5 were the result of analyzing them in the laboratory using a
portable spectroradiometer (FieldSpec). The reference values for jarosite were obtained from the USGS spectral
library [5], and the values were normalized and shown in Figure 5. Among the reflection characteristics of
jarosite, the absorption band seen near 900 nm is one of its characteristics [6]. The spectral characteristics of the
ground truth sample show slightly visible absorption bands. The spectrum curve obtained from the hyperspectral
image is unclear, although it can be seen depending on the appearance. More statistical analysis is needed in the
future, but in this study, we will limit the interpretation to this point. Alternatively, the sensitivity of the image
may be corrected in the future to improve the classification result. Therefore, the spatial distribution analysis of
the hyperspectral image was performed by the supervised classification performed by selecting the training data
from the Figure 6. The field survey selected the training data to determine the ROI range of trees, roads, and
rivers. For Jarosite M and Jarosite S, region of interesting (ROI) was created around the point where each ground
truth sample was collected (area surrounded by squares in Figure 6). The SAM method was used for supervised
classification.

Figure 5. Spectral features obtained from hyperspectral image and ground truth samples.

9th International Conference on Environmental Engineering, Science and Management
The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 411 -

Spatial distribution analysis results show that components with the same spectral characteristics as roads also
appeared on the surface of the sediment, and saying that it was successful is hard. However, the classification
of trees and water is well done, and the classification of Jarosite S and Jarosite M brings interesting results.
Regrettably, many roads were included, but Table 1 shows the control ratio in Figure 6 for each component.
From the distribution characteristics of Jarosite M and Jarosite S, the distribution of Jarosite M tends to be
closer to the river. This could suggest that the frequency of inundation affects the oxidation of surface
materials. Also, it seems that Jarosite M covers Jarosite S, and this could indicate that the more recently
deposited material contained more Jarosite M or that the sheet flood flow made the surface material more
oxidized. Furthermore, the planar structure of the deposit can also be estimated from this figure. In the area
close to the river channel, a lump-like, band-like, and sedimentary structure along the flow channel is seen,
whereas, in the area far from the river channel, it seems to exhibit a granular and individual distribution. The
surface sediment classified as the road represents that road materials are used in the surrounding area;
therefore, it is not so strange that the same materials are present. Also, in Figure 6, it seems to be distributed
far from the river channel, further below the jarosites M and S, and at the most downstream of the
meandering part of the river. This could be because sediment oxidation has not occurred in this area or the
acid streamflow has not reached this far. Nevertheless, the result of this analysis of hyperspectral images is
that the sediment containing jarosite covers most of the riverbank. Whether this will change over time will be
answered by reacquiring an image at a later time. Since it is UAV-based research, it could be a relatively
easy future, but we look forward to future research.

Figure 6. Spatial classification analysis result. The left side is the relevant part of the analysis in Figure 3.

Table 1. Distribution ratio of each component in Fig. 6
Image Class Samples Percent

Jarosite S 56,132 32.0

Jarosite M 48,945 27.9

Road 31,130 17.7

Water 5,371 3.10

Tree 33,826 19.3

Total 175,404 100

CONCLUSION
In this study, we used geoscience techniques and UAV techniques to investigate the state of the sediment in a
riverbank located downstream of a mine. Researchers found that the target deposit contained jarosite, a
secondary iron oxide mineral. Furthermore, an inspection of the sediment structure of the two soil pits and
chemical analysis of the sediment samples confirmed the presence of mining waste deposits in the riverbank.
Also, we estimated the area and amount of contaminated sediment containing these and obtained the
necessary numbers to remove these contaminations. Furthermore, the hyperspectral image obtained by UAV
confirmed that the pollutant was spreading over the entire ground surface.

9th International Conference on Environmental Engineering, Science and Management
The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 412 -

ACKNOWLEDGMENTS
This research was partially supported by the Japan Science and Technology Agency (JST) and the Japan
International Cooperation Agency (JICA) through Science and Technology Research Partnership for
Sustainable Development (SATREPS). We thank the principal investigator of the SATREPS, Prof. D.
Ishiyama and Dr. Z. Stevanović. This work was also supported by JSPS KAKENHI Grant Number
17K17613.
REFERENCES
[1] Macklin, M.G., Brewer, P.A., Hudson-Edwards, K.A., Bird, G., Coulthard, T.J., Dennis, I.A., Lechler,

P.J., Miller, J.R., and Turner, J.N. 2006. A geomorphological approach to the management of rivers
contaminated by metal mining. Geomorphology. 79: 423-447.
[2] Kurihara, J., Ishida, T., and Takahashi, Y. Unmanned Aerial Vehicle (UAV)-Based Hyperspectral
Imaging System for Precision Agriculture and Forest Management. In: Avtar, R. and Watanabe, T.
Unmanned Aerial Vehicle: Applications in Agriculture and Environment, Springer, 2020; 25-38.
[3] Nordstrom, D.K., Alpers, C.N. Geochemistry of Acid Mine Waters. In: Plumlee, G.S. and Logsdon,
M.J. eds., Chap. 6, The Environmental Geochemistry of Mineral Deposits, Rev. Econ. Geol. V. 6A,
Soc. Econ. Geol. Inc., Litteleton, CO, 1999.
[4] Japan International Cooperation Agency Economic Development Department (JICA report), 2008.
Actual situation and problems of mining activities: Mining Promotion Master Plan Survey Final
Report of the Republic of Serbia. 70-121.
[5] Kokaly, R.F., Clark, R.N., Swayze, G.A., Livo, K.E., Hoefen, T.M., Pearson, N.C., Wise, R.A.,
Benzel, W.M., Lowers, H.A., Driscoll, R.L., Klein, A.J., 2017, USGS Spectral Library Version 7: U.S.
Geological Survey Data Series 1035, 61 p., https://doi.org/10.3133/ds1035.
[6] Montero S., I. C., Brimhall, G. H., Alpers, C. N., & Swayze, G. A. 2005. Characterization of waste
rock associated with acid drainage at the Penn Mine, California, by ground-based visible to short-wave
infrared reflectance spectroscopy assisted by digital mapping. Chemical Geology, 215(1-4 SPEC.
ISS.): 453–472. https://doi.org/10.1016/j.chemgeo.2004.06.045

9th International Conference on Environmental Engineering, Science and Management
The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 413 - H-05

Carbon Footprint Assessment of Akha Mino Coffee Product

1Sirasit Meesiri and 2Anusorn Boopoke

1*Graduated student; 2 Assistant professor, Department of Environmental Engineering,
School of Energy and Environment, University of Phayao, 56000, Thailand.

*Phone : +66 89-992-8684, Fax : +66 5446-6704, E-mail : [email protected]

ABSTRACT
The greenhouse gas (GHG) emissions management at a product level is a key consideration for

many associated companies to mitigate global warming impacts. The carbon footprint of product (CFP) has a
concept to measure the product’s impacts on the environment in terms of CO2-equivalent (CO2eq), and assist
consumers identify alternative low-carbon products. In this study, the 100% Arabica M.Dark Roast 250 g.
was chosen to investigate carbon footprint calculation, which is a prominent product of Akha Mino Coffee
shop in Chiang rai province. The scope of this research aims to evaluate the carbon footprint beyond
business to customer (B2C). The data were collected in five stages during the producing process such as
extraction materials, production, distribution, utilization, and disposal. It was found that the total carbon
footprint (CFP) of 100% Arabica M.Dark Roast 250 g. was 2.46 kgCO2eq, in which the extraction materials
and disposal phases showed the significant amount of carbon footprint with 1.60 kgCO2eq (64.78%) and
0.63 kgCO2eq (25.70%), respectively. In contrast, the CFPs in stages of production and usage were
remarkably lower than others, which release 0.09 (3.50%) and 0.15 (6.01%) kgCO2eq, respectively.
According to the results, in order to reducing greenhouse gas emission, the manufacturers should provide
recommendation label to recycle the coffee grounds namely make organic fertilizers. Consequently, the CF
from disposal phases will be reduced effectively.

Keywords: life cycle assessment (LCA); carbon footprint of product (CFP); greenhouse gas; coffee product

INTRODUCTION
Coffee is the most popular consumed beverages because it is able to boost the energy, as well as

provide health benefits to reduce the risk of neurological diseases, cancer, type 2 diabetes and liver
dysfunctions [1]. It is considered as one of the significant income sources in Thailand’s agriculture sector
with the third ranking in the coffee production in Asia region [2]. There are two major coffee species in
Thailand including Arabica and Robusta, in which Arabica coffee is mainly planted in the North, specifically
in Chiang Rai and Chiang Mai [3]. It is because the Northern region is a mountainous area that can provide
ideal environments for Arabica plantation, with a spacious planting areas of more than 97x106 ha. and yield
approximately 800-850 tons per year [2].

Although Arabica coffee contributes well in the agricultural economics, it also causes considerably
detrimental impacts on the environment, especially greenhouse gas emission (GHG). It is illustrated that 1 kg
of roasted coffee from Nestlé company emitted 35 kgCO2eq [4]. Another publication in Germany reported
that the CFP of roasted coffee was 8.4 kgCO2eq kg−1, in which the stages of planting and on-site production
accounted more than a half of the total CFP, while the transportation, processing and waste disposal posse
15% of GHG emission [5]. Therefore, it is important to involve the coffee sector in measures of climate
change mitigation.

In order to have the overview of the product’s related environmental impacts as well as support
consumers in making environmentally responsible choices, it is necessary to provide relevant information
about them, especially it’s the carbon footprint of product (CFP) [6]. The CFP aims to supply informative
carbon dioxide (CO2) emission to consumers, so they can contribute towards reducing greenhouse gas
(GHG) emissions by purchasing low emission products, services, and enhance the competitiveness of the
Thai industrial sector in the global market. CFP is the calculation of carbon dioxide equivalent (CO2eq) of
the GHG emissions releasing from raw material acquisition, product manufacture, usage, waste management
and final disposal including related transports in all stages [7].

In this study, Arabica coffee from Akha Mino company was chosen to assess CFP. The Akha Mino
coffee is a prominent coffee shop in Chiang Rai (overview of the Akha Mino coffee shop was show in
Figure 1.). It produces agricultural-based products such as coffee beans, honey, rice, dried roselle and dried
chrysanthemum, etc. These products are not only sold in domestic markets, but also being exported to large

9th International Conference on Environmental Engineering, Science and Management
The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 414 -
markets namely Malaysia, Singapore, Indonesia and China, etc. This work aimed to evaluate GHG emissions
in the unit of CO2eq of 100% Arabica M.Dark Roast 250 g. as shown in Figure 2 in order to give useful
insights for the reduction GHG emissions from the coffee product.

(a) (b)

Figure 1 Overview of the Akha Mino coffee shop (a) and 100% Arabica M.Dark Roast 250 g. (b)

METHODOLOGY
The CFP was assessed by life cycle assessment (LCA) following the carbon footprint of product’s

guidebook. LCA is a standardized methodology, which provides reliable and transparent information and
being published by the International Organization for Standardization (ISO) in ISO 14040 and 14044 [8].
The description of four main phases of an LCA was shown in Figure 2.

Figure 2 The general phases of a life-cycle assessment
The assessment commences with determining goals and scopes. In this step, the GHG emission of
100% Arabica M.Dark Roast 250 g was evaluated in the form of CO2eq. Moreover, the boundary was
defined as Cradle to Grave or business to customer (B2C) as shown in Figure 3. Next is the inventory
analysis, in which 100% Arabica M.Dark Roast 250 g. inventory flows include inputs of materials and
supplies etc., and output consists of waste, wastewater and air pollution, etc. as shown in Figure 4. The third
step is the impact assessment (LCIA). The significance of potential environmental impacts of the product
based on the life-cycle impact flow results were assessed. In this study, we aimed at global warming impacts.
The last step is the interpretation. It is a systematic technique to identify, quantify, examine, and evaluate
information from the results of the life cycle inventory and/or the life cycle impact assessment of product.
9th International Conference on Environmental Engineering, Science and Management

The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 415 -

Figure 3 B2C boundaries of 100% Arabica M.Dark Roast 250 g. for CFP evaluation

Figure 4 The 100% Arabica M.Dark Roast 250 g. inventory analysis
9th International Conference on Environmental Engineering, Science and Management

The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 416 -
RESULTS AND DISCUSSIONS

The results of CO2 emission by LCA evaluation on global warming impact was 2.46 kgCO2eq of 100%
Arabica M.Dark Roast 250 g. as shown in Figure 5.

Fig 5. The results of 100% Arabica M.Dark Roast 250 g. CO2 emission on each sector
Figure 5 illustrates the CO2 emission on each sector relating life cycle of 100% Arabica M.Dark
Roast 250 g. The CO2 emissions were 1.60, 0.09, 0.00, 0.15 and 0.63 kgCO2eq of extraction of materials,
production, distribution, usage, and disposal, respectively. The analysis of CO2 emission sources is given by
the data on Figure 6.

Fig 6. CO2 emission percentage on each sector in the producing process of 100%
Arabica M.Dark Roast 250 g.

Figure 6 describes CO2 emission percentage of distinguished sectors in the manufactory process, in
which the stage of material extraction accounted for the highest amount of CO2 emission with 64.78%,
following is the disposal sector with 25.70%. However, in the stages of production, distribution and
utilization, CO2 emissions were small and not significant in the comparison with other stages.

Regarding the step of the extraction of materials with 64.78%, it involves with 99.23% of parchment
coffee, 0.73% of packaging and 0.04% of label. Added to this is CO2 emission from disposal stage is
25.70%, in which its include 99.96% of coffee grounds and 0.05% of packaging and label.

Therefore, in order to reduce CO2 emission, it is crucial to inform the data of CO2 emission. It is
because the share of the coffee grounds in the disposal step accounts the highest percentage. In other words,
the coffee grounds emitted the huge amount of CFP with 0.63 kgCO2eq (25.60% of CO2 emission of this

9th International Conference on Environmental Engineering, Science and Management
The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 417 -

product) in the comparison with all process (excepted extraction of materials because it was operated by

trader). As a result, if the organic residues from the product can be used in other purposes without

discharging into the landfill, the amount of CO2 emission will be cut down remarkably.
Finally, it is important to validate CFP results of Akha mino coffee product by comparing to other

Arabica’s CFP values from other coffee brands. The 500 g. of a Nex coffee and the 200 g. of a Doitung

espresso roast were used for this comparison with 250 g. of 100% Arabica M.Dark Roast, which are shown
in Table 1 and Figure 7. According to Table 1, CO2 emission of Akha Mino coffee product stays within a
range of two compared coffee products, in which the upper range is CFP value’s Doitung coffee and lower
range is from Nex coffee, with 5.00 kgCO2eq and 1.65 kgCO2eq, respectively.. The results show the CO2
emission of Doitung espresso roast product is higher 2.54 kgCO2eq in the comparison with the study product,
whereas the CFP value of Nex coffee is lower 0.81 kgCO2eq. The differences may come from distinguished
factors such as manufacturing process, distribution or disposal procedure. However, we could confirm the

CO2 emission value of 100% Arabica M.Dark Roast 250 g. was calculated reasonably when comparing with
the similar product type.

Table 1 The CFP comparison of the product by the similar type

Products CFP of comparison CFP of Akha Mino product Difference of CFP Ref.
(kgCO2eq) (kgCO2eq) (kgCO2eq)
Nex product [9]
Doitung product 1.65 2.46 +0.81 [10]
5.00 2.46 -2.54

(a) (b)
Fig 7. The 500 g. of a Nex coffee (a) and the 200 g. of a Doitung espresso roast (B)

CONCLUSION
The amount of CO2 emission by utilizing LCA in terms of global warming impact was 2.46

kgCO2eq of 100% Arabica M.Dark Roast 250 g. Moreover, the extraction of materials accounted for the
highest percentage with 64.78%, while the disposal stage ranked the second with 25.70%. Consequently, it is
recommended that in order to reduce CO2 emission, the CF value should be written on the product’s label, so
after being used, the organic residues can be solved by alternative ways instead of going to the landfill.

9th International Conference on Environmental Engineering, Science and Management
The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 418 -
REFERENCE
[1] Messina, G. , Zannella, C. , Monda, V. , Dato, A. , Liccardo, D. , De Blasio, S. , Valenzano, A. ,

Moscatelli, F. , Messina, A. , Cibelli, G. and Monda, M. 2015. The beneficial effects of coffee in
human nutrition. Biology and Medicine. 7(4).
[2] Angkasith, P. 2002. Coffee Production Status and Potential of Organic Arabica Coffee in Thailand.
Journal of Technology. 5(3).
[3] Noppakoonwong, U., Khomarwut, C., Hanthewee, M., Jarintorn, S., Hassarungsee, S., Meesook, S.,
Daoruang, C., Naka, P., Lertwatanakiat, S., Satayawut, K., Pereira, A. P., Silva, S. M. C., & Várzea,
V. M. P. (n.d.). Research and Development of Arabica Coffee in Thailand. 9.
[4] Nestle. 2010. Site Report e Himeji Factory. Nestle Group, Himeji, Japan.
[5] Tchibo. 2009. Case Study Tchibo Privat Kaffee Rarity Machare. PCF Pilot Project
Germany. Tchibo GmbH, Berlin, p. 55.
[6] Hassard, H.A. Couch, M.H. Techa-erawan, T. and McLellan, B.C. 2014. Product carbon footprint and
energy analysis of alternative coffee products in Japan. Journal of Cleaner Production. 73: 310-321.
[7] Thøgersen, J., and Nielsen, K.S. 2016. A better carbon footprint label. Journal of Cleaner Production.
125: 86-94.
[8] Thailand Greenhouse Gas Management Organization (TGO). 2018. Carbon footprint of product’s
guidebook. Thailand.
[9] Thailand Greenhouse Gas Management Organization (TGO), TGO CF 13-437. http://www.tgo.or
.th/2020/index.php/th/. May 19,2020.
[10] Thailand Greenhouse Gas Management Organization (TGO), TGO CF 15-129-491. http://www.tgo.or
.th/2020/index.php/th/. May 19,2020.

9th International Conference on Environmental Engineering, Science and Management
The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 419 - H-06

Fractionation of the main components of rice straw
by solvothermal process

Supawan Upajak1 and Saksit Imman1*

1 Master Student in Environmental Engineering, School of Energy and Environment, University of Phayao,
Tambon Maeka, Amphur Muang Phayao 56000, Thailand

1* Lecturer in Environmental Engineering, School of Energy and Environment, University of Phayao,
Tambon Maeka, Amphur Muang Phayao 56000, Thailand

*Phone : +66-86158-4845, Fax : 0-5446-6704, E-mail : [email protected]

ABSTRACT
The fractionation of rice straw using organosolv was performed under the acid concentration of 0.05M,
0.1M, and 0.2M of sulfuric acid as promoter for 30 min. The reaction temperature of 160°C was fixed all the
works. The optimal condition was focused on the highest cellulose yield in the isolated solid and maximum
yield of lignin detected in the organic phase. The results showed that pretreatment with 0.1M sulfuric acid
showed 95.8% of cellulose yield and highest hemicellulose solubilization higher than 90% from the native
rice straw. Scanning electron microscopy (SEM) revealed disruption of the intact biomass structure and
showed higher crystallinity index compared to the native biomass as shown by x-ray diffraction with a
marked increase in surface area as revealed by BET measurement. The work provides an insight into effects
of organosolv fractionation on modification of physicochemical properties of rice straw and an efficient
approach for its processing in biorefinery industry.

Keywords : Organosolv fractionation; Biomass; Rice straw; Biorefinery

INTRODUCTION
There are several procedures to minimize energy and environmental problems. One of the promising

current approaches is to develop clean alternative fuels to replace conventional oil. Lignocellulosic biomass
such as rice straw provides a low-cost feedstock for the biological production of fuels and chemicals, and
offers economic, environmental and strategic advantages. The dilute sulfuric acid pretreatment can
effectively solubilize hemicellulose into monomeric sugars and soluble oligomers, thus improving cellulose
conversion.

In this work, fractionation of rice straw using organosolv was performed under various concentration
of catalyzed condition. The work gives insights into the efficiencies of organosolv fractionation as an
alternative strategy for this potent starting material in biorefinery industry. Fractionation of rice straw using
organosolv was performed under acid promoter condition and its effects on physical and chemical
characteristics of the biomass were studied. In order to improve the efficiency of fractionation the rice straw
components to obtain high purity cellulose and low amount of by-product. Organosolv fractionation on
improve cellulose yield was investigated for utilization of cellulose that convert into glucose. The work gives
insights into the efficiencies of organosolv fractionation as an alternative strategy for this potent starting
material in biorefinery industry.

METHODOLOGY
Materials
Rice straw (RS) were obtained from local farmers of So village, Phayao, Thailand. The raw material was
dried at 60°C for 24 hr in the oven and sieved to particle size 1-2 millimeters using a cutting mill (Retsch
SM2000, Hann, Germany). The ground material was sieved using 1, 0.625 and 0.25mm sieves and respective
fractions were collected and stored in airtight containers at room temperature for further use. The chemicals
and reagents including sugar standards were purchased from Merck, Sigma-Aldrich. The biomass contained
35.8% cellulose, 21.5% hemicellulose, 24.4% lignin, and 15.0% ash on a dry-weight basis according to the
standard NREL method [1].

9th International Conference on Environmental Engineering, Science and Management
The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 420 -

Estimation of cellulose, hemicellulose and lignin
Cellulose and hemicellulose
The monomeric sugars (glucose, xylose, and arabinose) in the liquid fraction after reactions were analyzed
via HPLC for quantification of cellulose (glucose) and hemicellulose (xylose and arabinose). Initially, the
liquid fraction was neutralized with CaCO3 and filtered using 0.2μm filters for further analysis. HPLC (LDC
Model 4100, Shimadzu, Kyoto, Japan) analysis of filtered samples was performed using Aminex HPX-87H
column (Bio-Rad, Hercules, CA, USA) and 5 mM H2SO4 as mobile phase with a flow rate of 0.5 ml/min.
Oven temperature was maintained at 65 °C, and the refractive index detector was used. Glucose, xylose, and
arabinose were used as standards for HPLC analysis [2].

Lignin
The lignin content was measured as acid soluble and acid insoluble lignin. Lignin composition was
determined based on the Klason lignin content according to the method of laboratory analytical procedure
provided by the National Renewable Energy Laboratory (NREL) [1].

Organosolv fractionation of rice straw
The organosolv fractionation was performed reactions in Parr Reactor 4560. Rice straw residues was
conducted including of acid with three different H2SO4 concentration (0.05M, 0.1M, and 0,2M) in the close
system with pyrex glass reactor. The control experiments were carried out in DI water at the same condition.
Temperature inside the reactor was controlled in 160°C for 30 min, solid: liquid (S: L) ratio at 1g : 15ml in
Parr reactor. The initial pressure 20 bar was fixed in a temperature-controlled jacket with stirring at 100 rpm.
After reaction, the reactor was cooled in ice bath. The pretreated rice staw were separated by filtering on the
filter paper (Whatman No.4). The isolated solid was dried at 60°C for 24 hours and kept in a room
temperature. The liquid fraction was collected for analysis of sugar and inhibitory by-products by HPLC.

Analytical methods for lignocellulose components and products
The chemical compositions (% cellulose, hemicelluloses, lignin, and ash) in the solid and liquid phases were
determined following standard protocols given by National Renewable Energy Laboratory (NREL) [1].
Fermentable sugar profiles and inhibitory byproducts (5-hydroxymethylfurfural and furfural) were analyzed on a
high-performance liquid chromatograph (SPD-M10A DAD, Shimadzu, Kyoto, Japan) equipped with a refractive
index detector using an Aminex HPX-87H column (Bio-Rad, Hercules, CA, USA) operating at 65°C with 5 mM
H2SO4 as the mobile phase at a flow rate of 0.5 mL/min. The product yield is reported as the percentage of the
product obtained based on its content in the native rice straw on a dried weight basis. The reaction selectivity for
glucan (G), hemicellulose (H), and lignin (L) is calculated according to equations 2.1 and 2.2.

Product yield = ×100 (2.1)

Relative content = ×100 (2.2)

Physical analysis of solid residue
The microstructure of the native biomass and the solid residues obtained from the fractionation process was
analyzed by scanning electron microscopy (SEM) using a JSM-6301F scanning electron microscope (JSM-
6301F, JEOL, Japan). The samples were dried and coated with gold for analysis. An electron beam energy of
20 kV was used for analysis. The crystallinity of the native and separated solid fractions was determined by
X-ray diffraction (XRD) using an X'Pert PRO diffractometer (PANalytical, Almelo, The Netherlands). The
samples were scanned in a range of 2θ =10°–30° with a step size of 0.02° at 500 kV and 30 mA. The
crystallinity was calculated according to equation 2.4 [3].

CrI (%) = ×100 (2.4)

9th International Conference on Environmental Engineering, Science and Management
The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 421 -

Where I002 is the scattered intensity at the main peak of cellulose which typically lies around 002 plane at 2ϴ
= 22.4. And Iamorphous is the scattered intensity of amorphous portion evaluated as the minimum intensity
diffraction between the main and secondary (001 and 002 planes). (i.e. cellulose, hemicellulose, and lignin)
at 2ϴ = 18.0.
BET surface area measurement
The method of Brunauer, Emmett, and Teller (BET) was used to determine the total surface area of
materials. Raw and pretreated biomass samples were analyzed for the BET surface area using a Belsorp-max
TPDpro (BEL Japan, Tokyo, Japan) with thermal conductivity detector (Semi-diffusion type, 4-element W-
Re filament) at the National Nanotechnology Center, Thailand.
RESULTS AND DISCUSSIONS
Effect of temperature on the composition change of rice straw
The products yield fractionation is listed in Figure 1. After fractionation under all conditions used, the main
components of the solid phase were cellulose (71.4%) and lignin (38.5-59.8), while the hemicellulose
content reduce more than half compared to the native contents (20.5-42.3% in the solid found). This is
because cellulose is more compact structure, uniform and high crystalline structure compared to
heterogeneous hemicellulose and lignin (2). Hemicellulose solubilized into the liquid phase in the range of
35.2-36.6%. In the organic phase, lignin is the main products detected and no contamination of cellulose and
hemicellulose.

Figure 1 Effect of temperature on Ethyl acetate and Ketone fractionation A) products yield on solid phase B)
products yield in liquid phase C) products yield in organic phase

Effect of acid concentration on the composition change of rice straw
The effect of H2SO4 concentration on products yield is presented in Figure 2. The solid yields of isolated rice
straw decreased when increasing H2SO4 concentration at the same temperature. This is due to the dissolution
of hemicellulose and lignin during the treatment (3). Cellulose is the mainly content presented in the
pretreated residues. Hemicellulose in the solid phase was in the range of 52.0-71.4%. In the liquid phase,
increase H2SO4 concentration led to higher of degradation products of hemicellulose such as furfural in the
liquid phase. Lignin yield obtained in the organic phase of 54.8-81.9% was found with high purity more than
90%.

9th International Conference on Environmental Engineering, Science and Management
The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 422 -

Figure 2 Effect of H2SO4 concentration on Ethyl acetate and Ketone fractionation A) products yield on solid
phase B) products yield in liquid phase C) products yield in organic phase

Characterization of fractionated products
Scanning electron microscopy (SEM)
The physicochemical properties of the solid pulps were determined using SEM, BET, and XRD. The surface
of native rice straw was covered by a waxy layer and embedded matrix components (Figure 3 A1, A2). The
fractionation process led to disruption of the biomass surface and microstructure, as shown by the removal of
covered waxy materials together with the associated hemicellulose and lignin, resulting in highly pure
cellulose fibers as seen in the samples of fractionated solid (Figure 3B1, B2). The observation of distinct
cellulose fibers explains the greater digestibility of fractionated solid compared with the native biomass.
Crystallinity index of the solid fraction was examined using XRD. For all samples, broad peaks were
demonstrated at approximately 17° and 23.5°. The crystallinity of fractionated cellulose solid showed
decreased CrI (55.29%) compared with the native biomass (85.83%).

Figure 3 Scanning electron micrographs of native biomass (A1 and A2) and solid from fractionation under
the optimal conditions (B1 and B2)

9th International Conference on Environmental Engineering, Science and Management
The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 423 -

After organosolv fractionation at optimal condition, cavities and cracks in the plant cell wall were observed
in the pretreated solid, which reflect the removal of hemicellulose and modification of the surface lignin [4].
These structural modifications led to an increase in an available surface area related using BET analysis
(Table 1). Substantial increases in accessible surface area from 3.7 m2/g to 8.1 m2/g were observed after
organosolv fractionation. The accessible surface area is regarded as one of the most important factors
affecting the effectiveness of enzymatic digestibility [5]. The crystallinity of the isolated rice straw is also
considered an important factor influencing the efficiency of enzymatic hydrolysis. According to the analysis
of XRD patterns (Fig. 4), higher CrI (63.7%) was obtained for pretreated samples under the organosolv
fractionation optimal conditions compared with native rice straw (56.2%). The increases in CrI were
revealed to effects fractionation on the removal of the amorphous hemicellulose and lignin fractions in the
biomass, while it showed less effect on the disruption of the cellulose, which containing highly crystalline
structures. Increases in biomass crystallinity were reported in different biomasses pretreated by other
pretreatment technologies, e.g., Tamarix ramosissima pretreated under LHW pretreatment [6] and also other
pretreatment processes, e.g., microwave-assisted acid pretreatment [7] and steam explosion [8].

Table 1 BET surface area and crystallinity index of untreated and pretreated at optimal condition

Biomass Surface area (m2/g) Crystallinity index (%)

Native Isolated Native Isolated

Rice straw 3.7 8.1 56.2 63.7

Fig. 4 X-ray diffraction profiles of the native (-) and pretreated biomass (-)

CONCLUSION
In conclusion, organosolv fractionation of rice straw has been reported in this study. The results showed a
modification of the biomass structural and chemical properties of corncob during the reaction, which led to
remarkable enhance in cellulose yield and enhance delignification of lignin into organic phase. Moreover,
organosolv fractionation has been able to effectively isolated biomass, however, there are certain challengers
that need to be addressed such as high cost of organic solvent, difficulty in recycling and reuse. In this study
shows the potential for implementation of organosolv fractionation as an efficient fractionation method for
rice straw.

ACKNOWLEDGEMENT
This project was financially supported by research grants from the Thailand Research Fund and the Office of
the Higher Education Commission (MRG6080162) and Higher Education Research Promotion and
Thailand's National Research Universities (2559A31962002).

9th International Conference on Environmental Engineering, Science and Management
The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 424 -

REFERENCE
[1] Sluiter, A., Hames, B., Ruiz, R., Scarlata, C., Sluiter, J., Templeton, D., Crocker, D.:

Determination of structural carbohydrates and lignin in biomass. Laboratory analytical
procedure 1617, 1-16 (2008).
[2] Sindhu, R., Kuttiraja, M., Binod, P., Sukumaran, R.K., Pandey, A.: Physicochemical
characterization of alkali pretreated sugarcane tops and optimization of enzymatic
saccharification using response surface methodology. Renewable Energy 62, 362-368
(2014).
[3] Li, C., Knierim, B., Manisseri, C., Arora, R., Scheller, H.V., Auer, M., Vogel, K.P.,
Simmons, B.A., Singh, S.: Comparison of dilute acid and ionic liquid pretreatment of
switchgrass: biomass recalcitrance, delignification and enzymatic saccharification.
Bioresource technology 101(13), 4900-4906 (2010).
[4] Suriyachai, N., Champreda, V., Kraikul, N., Techanan, W., Laosiripojana, N.: Fractionation
of lignocellulosic biopolymers from sugarcane bagasse using formic acid-catalyzed
organosolv process. 3 Biotech 8(5), 221 (2018).
[5] Raita, M., Denchokepraguy, N., Champreda, V., Laosiripojana, N.: Effects of alkaline
catalysts on acetone-based organosolv pretreatment of rice straw. 3 Biotech 7(5), 340
(2017).
[6] Zhuang, X., Wang, W., Yu, Q., Qi, W., Wang, Q., Tan, X., Zhou, G., Yuan, Z.: Liquid hot
water pretreatment of lignocellulosic biomass for bioethanol production accompanying with
high valuable products. Bioresource technology 199, 68-75 (2016).
[7] Mikulski, D., Kłosowski, G.: Microwave-assisted dilute acid pretreatment in bioethanol
production from wheat and rye stillages. Biomass and Bioenergy 136, 105528 (2020).
[8] Marques, F.P., Silva, L.M.A., Lomonaco, D., de Freitas Rosa, M., Leitão, R.C.: Steam
explosion pretreatment to obtain eco-friendly building blocks from oil palm mesocarp fiber.
Industrial Crops and Products 143, 111907 (2020).

9th International Conference on Environmental Engineering, Science and Management
The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 425 - H-07

Short-Term Effects of Ambient PM2.5 from Biomass Burning on
Mortality from Respiratory Diseases during

the Smog Episode in Thailand

1Thao Ngoc Linh Nguyen, 2Sittichai Pimonsree, 2,3Patipat Vongruang,
4Pham Thi Bich Thao and 1Kritana Prueksakorn

1Graduate student, Faculty of Technology and Environment, Prince of Songkla University, Phuket 83000,
Thailand; 1Lecturer, Faculty of Technology and Environment, Prince of Songkla University, Phuket 83000,
Thailand; 2Assistant Professor, Atmospheric Pollution and Climate Change Research Unit, School of Energy and
Environment, University of Phayao, Phayao 56000, Thailand; 3Lecturer, Department of Environmental Health,
School of Medicine, University of Phayao, Phayao 56000, Thailand; 4Researcher, The Joint Graduate School of
Energy and Environment, King Mongkut’s University of Technology, Thonburi, Bangkok 10600, Thailand

*Phone: 054-466666, E-mail: [email protected]

ABSTRACT

Thailand is suffering a severe air pollution issue attributable biomass burning during a smog episode, in
which the highest Particulate Matter (PM) concentration level was recorded in March, 2012 in the period
from 2011 to 2017. Moreover, human health has been experienced significantly during the smog episode in
every year. Meanwhile, it also associates with economic burdens, mainly from the impacts of premature
mortality due to PM2.5 exposure. In this paper, we estimated the public health benefits in March, 2012 under
the scenario of without biomass burning (BB) emission, together with the associated economic costs by
utilizing BenMAP-CE program. The results show that when the biomass burning emission is eliminated, the
reduction of daily PM2.5 concentration ranges from 3.16 μg/m3 to 71.55 μg/m3 during March, 2012. From the
analysis, we quantified that the air quality enhancement could avoid 1,850 respiratory mortalities in the
18-99 age group. The associated economic benefit is estimated to be 20.89 billion Thai Baht (THB) due to
the avoidable respiratory mortalities. It is clear that biomass burning emission causes detrimental impacts on
many aspects, especially human health and economic costs. Therefore, this information gives insights into
the strong influences of BB emission for policy makers to establish realistic air pollution control strategies.

Keywords: health benefits assessment, biomass burning, air quality, BenMAP, respiratory mortality, PM2.5

INTRODUCTION
Air pollution is one of the most severe issues globally. According to World Health Organization

(WHO) statistics, more than 90% of the world population is living in areas where the air quality exceeds
WHO air quality standards, which causes 4.2 million deaths annually [1]. Among of the variety of emission
sources, BB emitted a huge amount of toxic pollutants, which causes detrimental impacts on ecosystem,
atmospheric chemistry and global warming [2]. The particulate matter problems frequently reach a peak
during March over the mainland of Southeast Asia [3]. Moreover, analysis in March during year 2011 to
2019 found that the observed monthly average PM10 concentration in the northern Thailand is incredibly
high in year 2012, in which year 2012 has the highest the number of hotspots. The monthly average of PM10
concentration was recorded as 139 µg/m3 in March, 2012 and the highest daily average was also reported in
March with 298 µg/m3 [4]. In addition, March was also confirmed as the peak month of BB emission in
Southeast Asia [5].

There are many evidences about the negative impacts of toxic air pollutants on human health. Firstly,
due to the exposure of particulate matter with aerodynamic diameters less than or equal to 2.5 micrometers
(PM2.5) during 1999-2014, there were 1,256,300 premature deaths in South and Southeast Asia, in which
stroke and ischemic heart disease were the two main contributors in South Asia countries, especially India,
Bangladesh and Pakistan [6]. Second, the inhalation dose was investigated to estimate the impacts of PM2.5
on the human health in Chiang Mai, Thailand. To be more specific, the means estimated inhalation dose for
PM2.5 were 1,533 µg and 1,470 µg in Doi Ang Khang and Chiang Mai university, respectively [7].
Moreover, during 1999-2006, years with strong impacts attributable El Niño phenomenon shown the
remarkable contribution of landscape fires with 200 µg/m3 in PM2.5 annual average concentration in

9th International Conference on Environmental Engineering, Science and Management
The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 426 -

Southeast Asia. As a result, the annual increase of cardiovascular mortality in adults was estimated with
10,800 (6,800-14,300) persons [8]. These harmful influences are caused by PM2.5’s extremely small size,
hence it is easy to transport toxic compositions through the hair nose’s filtration, and assemble into the end
of respiratory tract, which will destruct other body organs because the air way in lungs is blocked [9,10,11].

The air pollution affects not only the human health, ecosystem but also the economic. According to
the Cost of Air pollution report of World Bank, in MSEA region, total welfare losses of Laos, Thailand and
Vietnam in 2013 were $2,409 (7.63% GDP equivalent), $63,369 (4.07% GDP equivalent) and $23,832
million (4.80% GDP equivalent), respectively [12]. Additionally, according to Organization for Economic
Cooperation and Development (OECD) report, if the unlikely situations of serious air pollution are still
remained, the outdoor air pollution will cost 1% of global GDP by 2060 with around USD 2.6 trillion
annual [13].

Among various methods to appraise impacts of air pollution in terms of human health and economic,
the Environmental Benefits Mapping and Analysis Program-Community Edition (BenMAP-CE) is a well-
established damage-function approach in recent years [14]. It is integrated into Geographic information
system (GIS) tool, which performs health impacts assessment and economic benefits from alterations in air
quality [15]. BenMAP is a Window based-program, which is well-established by U.S Environmental
protection agency (EPA), and being one of the most useful tools in the field of air pollution, which gives the
output in grid scales. BenMAP was applied widely, especially in United States, in many aspects such as the
benefit assessment of using electric vehicles [16], the benefits of reducing the maternal exposure to
particulate matters [17], the impacts of aviation biofuel supply [18] or a tool to provide useful information in
the policy decision-making [19]. There are several BenMAP-based papers in Asia [14,20,21-25], which are
mostly from China and Korea, and there is one paper in Vietnam in Southeast Asia region [26]. However,
input variables from Vietnam were extracted from US database, which show the high differences about the
nutritional characteristics and income levels between a SEA country and US [27]. Added to this is that the
health benefits assessment in SEA region is rare with only a paper about the effects of coal plant emission in
terms of premature deaths [28].

According to the monitoring data from Thailand Pollution Control Department, observed PM2.5 were
much higher than the national ambient air quality standard (daily average PM2.5 < 50 μg/m3) in year 2019.
Consequently, PM2.5 is a predominant issue in Thailand and draw high attention to citizens. Although there
are many studies about the characteristics of BB, its chemical compositions, BB inventories and impact of
BB on ambient concentrations, there is limited number of studies about the health benefits and economic
benefits of changing PM2.5 due to BB reduction in Southeast Asia region, especially in Thailand. These
issues were examined in this work.

METHODOLOGY
3.1 Health benefit’s estimation

In order to appraise benefits of the improvement of air quality attributable to no biomass burning, the
Environmental Benefits Mapping and Analysis Program-Community Edition (BenMAP-CE), version 1.5
was applied. It is a well-established damage-function approach in recent years [14]. BenMAP is known as
the program to estimate the human health impacts and associated economic benefits of the alteration of air
quality. Moreover, BenMAP assists to compare the changes of before and after implementing the
governmental policies or evaluate the potential of better air quality-based projects. BenMAP requires
multiple input datasets in the format of grid cells following the log-linear health effect function, which is
available in BenMAP set up,

(1),

where is denoted as the incidence change when air quality is changed positively, is the concentration-

response function (CRF), is the changes in individual air pollutant’s concentration, is the baseline-

incidence rate, Pop is the exposed population in the year same with the air quality is recorded. The output is

performed in terms of point estimates per grid cell and GIS maps. Therefore, each input dataset is required to

be formatted in grid cells, except .

9th International Conference on Environmental Engineering, Science and Management
The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 427 -

In this research, the is the difference of PM2.5 concentrations in the baseline case and the

control case. The baseline is the PM2.5 concentration accounting all of the emission sources in March, 2012,
whereas the control case is the simulation without the biomass burning emission. The dataset of PM2.5
concentration was simulated by Community Multiscale Air Quality Modeling System (CMAQ) with the grid

resolution 27 km 27 km [4]. Next, the was selected from the epidemiological study with systematic

review and meta-analysis in the scale of low-income and middle-income countries, which recorded the

relative risk (RR) between the PM2.5 short-term exposure and the daily mortality. The RR was 0.57%

(0.28-0.86) in the association between the increase of 10 3 PM2.5 in the same day and the increase in

respiratory mortality [9]. The ages of target population in this epidemiological study was from 18 to 99 for

both genders. The was extracted from the dataset of Global Burden of Disease Collaborative Network

(GBD), which included the death rates of both gender aged 18 to 99 in terms of related respiratory

causes [29]. These causes were coded as J00-J99 in the international Statistical Classification of Diseases

and Related Health Problems (ICD), version 10, which was in the corresponding the target causes from the

mentioned epidemiological study [30]. Moreover, the gridded population 2010 was taken from the Center for

International Earth Science Information Network (CIESIN), version 4 [31]. It was estimated to the Thai

population in 2012 and matched the registration data in 2012 from Department of Provincial Administration,

Ministry of Interior. These datasets should be performed in the grid scale before inputting into BenMAP. As

a result, the point estimates of health incidences were calculated and deliberated the result in grids. The

conceptual framework of this study is shown in Figure 1.

Fig.1 The Conceptual Framework of this study

3.2 Associated economic benefits
The monetized benefits were calculated by the value of statistical life (VSL), which was defined as

the meant of the aggregation of the individual valuation from Willingness To Pay (WTP) surveys. In another
way, the VSL value is a mean statistical value that a group of person willing to pay to marginally reduce
their risk of premature death [24]. In this research, the mean VSL was from OECD. The VSL is calculated
using equation (2):

(2),

where WTP is the amount of payment to reduce the risk of death, is the magnitude of risk reduction.

9th International Conference on Environmental Engineering, Science and Management
The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 428 -
Due to the mean VSL value derived from OECD, the benefit transfer methodology was applied with
the purpose of calculating the VSL of Thailand in USD, 2012. The benefit transfer method is method that
utilizes the estimated value from existing studies, and adjusts these values to estimate the amount of money
in the research-based area [8]. The benefit transfer’s equation is shown below,

(3),
where ℎ ,2012 is the VSL value for Thailand in 2012 Thai baht at 2012 Thai income levels.
,2005 is the OECD base value of 3 million in year 2005 US dollars at 2005 OECD income levels [25].
is the per capita GDP of the specified country in the specified year, expressed in constant international
(PPP-adjusted) dollars. is the income elasticity of the VSL, and OECD recommended a value with 1.2 in
the range from 1.0 to 1.4 for low, middle-income countries. 2005 is the Purchasing Power Parity index in
2005 in units of Thai baht per international dollar. ℎ d is the consumer price index in Thailand in the
specified year. These applied economic indicators in the equation (3) were taken from the database of World
Development Indicators, World Bank [32].

As a result, the VSL of Thailand was estimated and multiplied with the number of avoided deaths
from the previous step, in order to obtain the associated economic benefits.
RESULTS AND DISCUSSIONS

The results shown significant benefits in the scenario of without the emission of biomass burning. The
avoided respiratory mortality could gain 1,850 cases (95% CI: 890-2,783) for people ages 18 to 99 years
under the alteration of daily PM2.5 ranges from 3.16 μg/m3 to 71.55 μg/m3 during the smog episode in March,
2012 (Figure 2). Regarding the associated economic benefits, the values of the avoided respiratory mortality
was 20.89 billion Thai Baht (THB) (95% CI: 10.16-31.44 billion THB) as shown in Figure 2.

Fig.2 Number of avoidable respiratory deaths (left) and economic benefits from avoidable Respiratory
deaths (million THB/(grid.month) ) (right)

9th International Conference on Environmental Engineering, Science and Management
The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 429 -

There are high spatial variations of public health benefits associated with two principle factors: the
change in PM2.5 concentrations from biomass burning reduction and the number of populations. The results
indicated the critical benefits in central region and the Northern of Thailand. To be more specific, the central
region includes Bangkok metropolis and other surrounding provinces namely Pathum Thani, Samut Prakan
and Nakhon Pathom. Although the changing of PM2.5 concentration in these areas are lower than that of the
Northern of Thailand, the larger benefits in Bangkok and surrounding areas are because of the populated
areas. The impacts of the distribution of population in this benefit assessment are shown clearly in Figure 3,
which there are high population in Bangkok and surrounding provinces.

The highest public health benefits in Northern Thailand found in Chiang Mai province, particularly
the urban area with outstanding number of avoided respiratory deaths and associated economic benefits.
Added to this is the essential benefits spread out across Northern provinces. It is because although the
population of individual provinces is lower than Bangkok, the reduction of PM2.5 concentration attributable
biomass burning emission is remarkable. Ranking the second position of potential benefits in the North is
Chiang Rai province, following is Nan province. It is noticed that the associated benefits are widely
distributed to almost Northern Thailand including Phayao, Lampang, Lamphun, Tak, Sukkhothai and
Kamphaeng Phet.

Fig.4 The changes of PM2.5 concentrations in case of without BB emission (μg/m3) (left) and the
distribution of Thai population in 2012 (right)

The assumptions and assessments may have uncertainties. First, when CRF was applied in this work,
we assumed that there is no difference in the composition of PM2.5 from various emission sources, and the
Thai citizens have the same risk of mortality under the same PM2.5 exposure. Second, there are no country
specific values for health assesment calculation, including CRF, baseline incidence rates, and gridded
population dataset. Third, even though the simulated PM2.5 data are good in terms of spatial coverage.
However, they may have uncertainty in air quality modeling system. Last, although VSL is used widely to
estimate the public health benefits, especially in areas that have limited Willingess to pay’s statistics, there
are uncertainties in the theory and empirical aspects.

CONCLUSION
Results from public health benefits emphasized the biomass burning emission causes the short-term

impacts on human health during the smog episode in Thailand. Under the scenario of the impact of biomass
burning in March 2012, it showed that BB increased the daily PM2.5 concentrations in the range from 3.16
μg/m3 to 71.55 μg/m3, it could cause 1,850 respiratory mortalities for people in the 18-99 age group. As a
result, the associated economic losses were 20.89 billion THB due to respiratory mortalites in March, 2012.

9th International Conference on Environmental Engineering, Science and Management
The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 430 -

The highest health benefit of decrease in BB was found in Bangkok and suround provinces in association
with the most populated areas in Thailand and PM2.5 change moderately. Moreover, there were high health
benefits in Northern Thailand due to the largest decrease in the daily mean PM2.5 concentration when BB
was controls. The results of this study show clear health benefits and theirs economic benefits of decreasing
in BB. The information should be use for decission support to control BB in Thailand and surrounding area.
Additionally, high health impact area from PM2.5 should be focus to take care of people health.

ACKNOWLEDGEMENT
This work was supported by Prince of Songkla’s Graduate school financial support for the thesis

fiscal year 2019. We thank for the assistances of members in Atmospheric Pollution and Climate Change
Research Unit, School of Energy and Environment, University of Phayao and BenMAP support team from
US EPA.

REFERENCE
[1] WHO | Air pollution. Retrieved April 7, 2020 from http://www9.who.int/airpollution/en/
[2] P. J. Crutzen and M. O. Andreae. 1990. Biomass burning in the tropics: impact on atmospheric

chemistry and biogeochemical cycles. Science. 250 (4988):1669–1678.
[3] Sittichai Pimonsree and Patipat Vongruang. 2018. Impact of biomass burning and its control on

particulate matter over a city in mainland Southeast Asia during a smog episode. Atmospheric
Environment. 195:196–209.
[4] Patipat Vongruang and Sittichai Pimonsree. 2020. Biomass burning sources and their contributions to
PM10 concentrations over countries in mainland Southeast Asia during a smog episode. Atmospheric
Environment. 228: 117414.
[5] Libo Zhang, Yongqiang Liu, and Lu Hao. 2016. Contributions of open crop straw burning emissions to
PM 2.5 concentrations in China. Environtal Research Letter. 11 (1): 014014.
[6] Yusheng Shi, Aimei Zhao, Tsuneo Matsunaga, Yasushi Yamaguchi, Shuying Zang, Zhengqiang Li,
Tao Yu, and Xingfa Gu. 2018. Underlying causes of PM2.5-induced premature mortality and potential
health benefits of air pollution control in South and Southeast Asia from 1999 to 2014. Environment
International. 12: 814–823.
[7] Shantanu Kumar Pani, Somporn Chantara, Chanakarn Khamkaew, Chung-Te Lee, and Neng-Huei Lin.
2019. Biomass burning in the northern peninsular Southeast Asia: Aerosol chemical profile and
potential exposure. Atmospheric Research. 224: 180–195.
[8] Miriam E. Marlier, Ruth S. DeFries, Apostolos Voulgarakis, Patrick L. Kinney, James T. Randerson,
Drew T. Shindell, Yang Chen, and Greg Faluvegi. 2013. El Niño and health risks from landscape fire
emissions in southeast Asia. Nature Climate Change. 3 (2): 131–136.
[9] Katherine Newell, Christiana Kartsonaki, Kin Bong Hubert Lam, and Om P Kurmi. 2017.
Cardiorespiratory health effects of particulate ambient air pollution exposure in low-income and
middle-income countries: a systematic review and meta-analysis. The Lancet Planetary Health. 1 (9):
e368–e380.
[10] Yu-Fei Xing, Yue-Hua Xu, Min-Hua Shi, and Yi-Xin Lian. 2016. The impact of PM2.5 on the human
respiratory system. Journal of Thoracic Disease. 8 (1): E69–E74.
[11] Yongquan Yu, Shen Yao, Huibin Dong, Li Wang, Chao Wang, Xiaoming Ji, Minghui Ji, Xingjuan
Yao, and Zhan Zhang. 2019. Association between short-term exposure to particulate matter air
pollution and cause-specific mortality in Changzhou, China. Environmental Research. 170: 7–15.
[12] The Cost of Air Pollution | READ online. Retrieved March 29, 2020 from https://read.oecd-
ilibrary.org/environment/the-cost-of-air-pollution_9789264210448-
en?fbclid=IwAR1LuvJIEgtfrubJobBpN4RKwB6pmuJTa8aTTo301mbbbu8vbEZYrVVhZ2s#page54
[13] The Rising Cost of Ambient Air Pollution thus Far in the 21st Century: Results from the BRIICS and
the OECD Countries. 2017.
[14] Richard A. Broome, Neal Fann, Tina J. Navin Cristina, Charles Fulcher, Hiep Duc, and Geoffrey G.
Morgan. 2015. The health benefits of reducing air pollution in Sydney, Australia. Environmental
Research. 143: 19–25.

9th International Conference on Environmental Engineering, Science and Management
The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 431 -

[15] Jason D. Sacks, Jennifer M. Lloyd, Yun Zhu, Jim Anderton, Carey J. Jang, Bryan Hubbell, and Neal
Fann. 2018. The Environmental Benefits Mapping and Analysis Program – Community Edition
(BenMAP–CE): A tool to estimate the health and economic benefits of reducing air pollution.
Environmental Modelling & Software. 104: 118–129.

[16] Shuai Pan, Anirban Roy, Yunsoo Choi, Ebrahim Eslami, Stephanie Thomas, Xiangyu Jiang, and H.
Oliver Gao. 2019. Potential impacts of electric vehicles on air quality and health endpoints in the
Greater Houston Area in 2040. Atmospheric Environment. 207: 38–51.

[17] Jina J. Kim, Daniel A. Axelrad, and Chris Dockins. 2019. Preterm birth and economic benefits of
reduced maternal exposure to fine particulate matter. Environmental Research. 170: 178–186.

[18] Vikram Ravi, Allan H. Gao, Natalie B. Martinkus, Michael P. Wolcott, and Brian K. Lamb. 2018. Air
quality and health impacts of an aviation biofuel supply chain using forest residue in the northwestern
united states. Environmental Science and Technology. 52 (7): 4154–4162.

[19] Jean E. Johnson, David L. Bael, Jeannette M. Sample, Paula G. Lindgren, and Dorian L. Kvale. 2017.
Estimating the Public Health Impact of Air Pollution for Informing Policy in the Twin Cities: A
Minnesota Tracking Collaboration. Journal of Public Health Management and Practice. 23: S45–S52.

[20] Hyun Joo Bae and Jeongim Park. 2009. Health benefits of improving air quality in the rapidly aging
Korean society. Science of The Total Environment. 407 (23): 5971–5977.

[21] Li Chen, Mengshuang Shi, Suhuan Li, Shuang Gao, Hui Zhang, Yanling Sun, Jian Mao, Zhipeng Bai,
Zhongliang Wang, and Jiang Zhou. 2017. Quantifying public health benefits of environmental strategy
of PM2.5 air quality management in Beijing–Tianjin–Hebei region, China. Journal of Environmental
Sciences. 57: 33–40.

[22] Dian Ding, Yun Zhu, Carey Jang, Che-Jen Lin, Shuxiao Wang, Joshua Fu, Jian Gao, Shuang Deng,
Junping Xie, and Xuezhen Qiu. 2016. Evaluation of health benefit using BenMAP-CE with an
integrated scheme of model and monitor data during Guangzhou Asian Games. Journal of
Environmental Sciences. 42: 9–18.

[23] Jongsik Ha and Nankyoung Moon. 2013. Uncertainty and Estimation of Health Burden from
Particulate Matter in Seoul Metropolitan Region. Journal of Korean Society for Atmospheric
Environment. 29 (3): 275–286.

[24] Daeun Kim, Jeongyeong Kim, Jaehwan Jeong, and Minha Choi. 2019. Estimation of health benefits
from air quality improvement using the MODIS AOD dataset in Seoul, Korea. Environmental
Research. 173: 452–461.

[25] Jiabin Li, Yun Zhu, James T. Kelly, Carey J. Jang, Shuxiao Wang, Adel Hanna, Jia Xing, Che-Jen Lin,
Shicheng Long, and Lian Yu. 2019. Health benefit assessment of PM2.5 reduction in Pearl River Delta
region of China using a model-monitor data fusion approach. Journal of Environmental Management.
233: 489–498.

[26] Bang Quoc Ho. 2017. Modeling PM10 in Ho Chi Minh City, Vietnam and evaluation of its impact on
human health. Sustainable Environment Research. 27 (2): 95–102.

[27] Achilleos, S., Kioumourtzoglou, M.-A., Wu, C.-D., Schwartz, J. D., Koutrakis, P., & Papatheodorou,
S. I. 2017. Acute effects of fine particulate matter constituents on mortality: A systematic review and
meta-regression analysis. Environment International. 109: 89–100.

[28] Shannon N. Koplitz, Daniel J. Jacob, Melissa P. Sulprizio, Lauri Myllyvirta, and Colleen Reid. 2017.
Burden of Disease from Rising Coal-Fired Power Plant Emissions in Southeast Asia. Environmental
Science and Technology. 51 (3): 1467–1476.

[29] Global Burden of Disease Collaborative Network. Global Burden of Disease Study 2017 (GBD 2017)
Results. Seattle, United States: Institute for Health Metrics and Evaluation (IHME), 2018. Available
from http://ghdx.healthdata.org/gbd-results-tool.

[30] ICD-10 Version:2019. (n.d.). Retrieved April 8, 2020, from https://icd.who.int/browse10/2019/en
[31] Center for International Earth Science Information Network - CIESIN - Columbia University. 2018.

Gridded Population of the World, Version 4 (GPWv4): Administrative Unit Center Points with
Population Estimates, Revision 11. NASA Socioeconomic Data and Applications Center (SEDAC).
[32] World Development Indicators | DataBank. Retrieved March 30, 2020 from
https://databank.worldbank.org/reports.aspx?source=world-development-
indicators#advancedDownloadOptions

9th International Conference on Environmental Engineering, Science and Management
The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 432 -

H-08

Design of Green Roof Application with Gray Water in Tropical Area

Moritani Shigeoki1* Tahei Yamamoto2 Sukthai Pongpattansiri3
Chuleemas Boonthai Iwai4 and Anusorn Bunpok3

1Faculty of Agriculture and Life Science, Hirosaki University, 3 Bunkyo, Hirosaki, Aomori, Japan;
2Arid Land Research Center, Tottori University, Hamasaka 1390, Tottori, Japan; 3The School of Energy

and Environment, University of Phayao, Phayao, 56000, Thailand; 4Faculty of Agriculture,

Khon Kaen University, Khon Kaen, 40002. Thailand
*Phone: +81-172-39-3849, E-mail: [email protected]

ABSTRACT
While maintaining the greenery of cities is considered the major function of a green roof, it could also be
helpful in using gray water during water scarcity and for constructing bee habitats for commercial uses such
as honey production and education. In this study, gray water was simulated by saline irrigation. It had an
electrical conductivity (EC) of 18 dS m-1 and was applied for irrigation to determine the stress factors
incurred by three kinds of plants, a turf grass (Cynodon dactylon) and two Crassulacean acid metabolism
(CAM) plant species, Sedum kamtschaticum and Sedum oryzifolium. The stress factor was defined as time
integration with the soil water potential beyond the criteria for normal plant growth. Plants showed decreased
evapotranspiration and increased water use efficiency based on stress factor increments; whether the rates
increased or decreased depended on the plant species. The cations Na+, Ca2+, Mg2+, and K+ accumulated in
leaves based on the irrigation salinity, which was proportional to the EC of leaves. Using the EC of drainage
water as the representative soil EC measurement seems potentially inaccurate owing to the thin growing
medium, and the results could easily vary based on the position of the soil profile and timing of the
measurement. Sampling leaves and measuring the EC was found to be an accurate way to assess plant health.
A better understanding of the influence of stress factors on evapotranspiration could help in selecting suitable
plants for an area, based on the soil salinity and the potential for drought.

Keywords: bee habitats, soil water potential, evapotranspiration, intensive green roof in tropical area, stress
factor, saving of water requirement

9th International Conference on Environmental Engineering, Science and Management
The Heritage Chiang Rai, Thailand, May 27-29, 2020

- 433 -

INTRODUCTION
Agriculture on urban rooftops has received attention recently
because it provides local nutritious foods to urban communities
as well as a site for social events that have an educational impact

Figure 1. Schematic layout of gray water irrigation for

vegetation with beehives (Left) and thin substrate layers

of a representative green roof system (Right). Figure 2. Schematic layout of gray water

on multiple generations [1]. Awareness of greenery in urban collection and usage for irrigation on a
areas is being raised in order to maintain ecosystem diversity rooftop after being lifted by a pump into
and prevent biological extinction, as there have been reports of the primary tank, based on Friedler and
declining wild bee species for a few decades, due to habit loss Hadari, 2006 [7] and Agra et al., 2018 [8].

and pesticides [2]. Bees have an important role as pollinators of

most flowering plants, helping to maintain urban ecosystems in parks and other green spaces. Bee diversity

and syrphid visits are significantly affected by plant diversity, especially floral abundance [3]. The warmer

climate of cities is preferable for many wild bees due to their ectothermic and thermophilic nature [4]. Thus,

green rooftops in tropical areas could provide alternative habitats for bees under carefully managed
conditions such as the intensive green roof and under preferable design of roof configuration for bees’ visits

with beekeepers shown in Figure 1 [5]. Bees create natural honey that can be harvested from the green roof

[6]; this can serve as an educational experience for students as well as a source of income for businesses

selling jars of honey. However, the additional demand for irrigation water beyond that used for daily living

has to be taken into account when considering the intensive green roof.

Increasing population growth in urban areas leads to water scarcity, but the treated wastewater (gray water)
originating from water used for sinks, baths, clothes and washing machines as well as collected rainfall can
be utilized for toilet flushing (Figure 2). The gray water helps considerably reduce the urban water demand,
although its uses are limited due to the possible health risks from any remaining pollution [7]. In order to
apply gray water to a green roof, complicated processes, such as removing organic matter indicated by the
biological oxygen demand (BOD) and suspended solids (SS) and adjusting the chemical composition for
optimal plant growth, would be inevitable [8]. However, the removal of inorganic components such as Na+
seems difficult due to the maintenance cost of ion exchange filtration systems. Therefore, when gray water is
used for green roof irrigation, saline water would also need to be applied to vegetation to some extent,
depending on the mixture of tapped water and collected rainfall. Thus, plants need to be selected based on
their tolerance for both salinity and drought; this is especially true due to the possibility of a long duration of
dry climate and high water demand from residents, industrial and field agricultural activities, limiting the
water available to irrigate the green roof [9].

9th International Conference on Environmental Engineering, Science and Management
The Heritage Chiang Rai, Thailand, May 27-29, 2020