BIOLOGICAL MONITORING OF SOLAR UV RADIATION.

BIOLOGICAL MONITORING

OF SOLAR

UV

RADIATION

AT

17

SITES

IN

ASIA,

EUROPE

AND

SOUTH

AMERICA FROM

1999

TO

2OO4

Noeuo

MuNaKarR, SaNroso

CoRnnrN,

Mpu

Keruoro,

Ketur

MuLyRot,

Snt LesraRI,

Wroooo

WrRoHaoro.lo;o, DRvro

80lSee,

Srelros

KezaDzrs,

Vrcron

Meven-Rocnow,

NenoN

ScHUcH,

Cmuoro Cesrccn,

MoroHtse KaNEro, CHuNc-MtNc LIu,

KowtcHt

Jtivleow, TosHrat<r SRrDe, CHrrnrco NtsHtcoRt,

Knrsuvtr

OcnTe,

Kazunrno

INRFUKU,

Koreno

HrEon eNo

Mesautrsu

IcHrunsut

Made in United States of America

Repri nted from PuorocuEMrsrR y A:{ D Paorostor-ocv Vol. 82, No. 3. May/june 2006

Biological Monitoring

of

Solar

UV Radiation

al

17

Sites in

Asia,

Europe and South America

from

1999

to

2004

Nobuo

Munakata*1,

Santoso Cornain2, Mpu Kanoko2, Ketut Mulyadi3,

Sri

Lestaria,

Widodo Wirohadidjojos, David Bolse66,

dtetios

Kazadzis,T-,

Victor

Meyer-Rochowt't,

Nelson Schuchlo,

|laudio

Casicciall,

Motohisa

Kaneko12,

Chung-Ming

Liu13,

Kowichi Jimbowla, Toshiaki

Saida15,

Chikako Nishigoril6, Katsumi Ogatal7,

Kazuhiro lnafukulE, Kotaro Hiedal and Masamitsu lchihashil6

lBiophysics _{Laboratory and Frontier Project "Life's Adaptation Strategies}

to

_{Environmental Changes,"} Faculty

oi

Science, Rikkyo University, Tokyo, Japan

2Depariment _{of Anatomic Pathology, University of lndonesia,} Jakarta, lndonesia 3Department _{of Anatomic Pathology, Udayana University,} Denpasar, lndonesia aDermatology _{and Venereology Department, Andalas University, Padang, lndonesia} sDepartmenf _{of Dermatology, Gadjah Mada University, Yogyakarta, lndonesia} 6Belgian lnstitute of Space Aeronomy, Brussels, Belgium

tphyiics

_{Department, Aristotle University of Thessaloniki, Thessaloniki, Greece}

sDepartment _{of Biology, University of Oulu, Oulu, Finland} eSchool

of

_{Enqineerinq and Science, lUB, Bremen, Germany}

losouthem Re!ional Space Research Centre (CRSPE/lNPE-MCT), Santa Maria, Brazil

ttOzone

_{and UV Radiation Laboratory, University of Magallanes, Punta} Arenas, Chile

t2Department _{of Biomedical Engineering, Jordan University of Science and Technology, lrbid, Jordan} l3Department _{of Atmospheric Sciences, National Taiwan University, Taipei, China}

laDepartment _{of Dermatology, Sapporo Medical University, Sapporo, Japan}

lsDepartment _{of Dermatology, Shinshu University School of Medicine, Matsumoto, Japan} l6Department _{of Dermatology, Kobe University, Kobe, Japan}

lTDepartment _{of Dermatology, Miyazaki University, Kiyotake, Miyazaki' Japan}

lsDivision of Dermatology, D'epartment of Organ Oriented Medicine, University of the Ryukyus, Nishihara, Okinawa, Japan

Received 7 July 2005; accepted 4 November 2005; published online 8 November 2005 DOI: 10.15622005-07-07-RA-602 Photochemistry and Photobiology, 2006,

82:

689-694

ABSTRACT

A small and robust dosimeter for determining the biologically effective dose

of

ambient

UV

radiation

has been developed using LlV-sensitive mutant spores

of

Bacillus

snlrrlis

strain

TKI6f

2. A membrane

filter

with four spots of the spores was snapped to a slide mount. The slide was wrapped and covered

with

two

or

more layers

of

polyethylene sheet

to

protect the sample from rain and snow and to reduce monthly-cumulative doses

witNn

the measurable range. From 1999, monthly data were collected

at

17 sites

for

more than 1 year, and data

for

4 to 6 consecutive years were obtained

from

12 sites. Yearly total values of the spore inactivation dose

iSID)

ranged from 3200

at

subarctic

Oulu

to

96000

at

tropical

Denpasar, and

the

mean

yearly

values

of

SID

exhibited

an

exponential dependence on latitude

in

both hemispheres

with

a doubling

for

about every 14 degrees of change. During the observation period, increasing trends of

UV

doses have been observed at" all sites with more than 5 years of data available. Year'to-year variations

at

high

and

middle latitude

sites

are

considered

tCorresponding author email: nobmunak@ric.rikliyo.ne.jp (Nobuo Munakata)

O ?006 Americm Society for Photobiology 0031-8655/06

due

mostly

to

climatic

variation.

At

three tropical

sites' negative correlations betrveen the yearly doses and the colum-n ozone amounts were observed. The results verified the

appli-cability

of

spore dosimetry

for

global

and

long-time

moni-toring

of

solar

UV

radiation,

in

particular

at

tropical

sites where no monitoring is taking place.

INTRODUCTION

Solar

tIV

radiation is a ubiquitous environmental agent affecting many terrestrial and aquatic

life

forms. Harmful effects

of de

exposure are incurred mainty by the induction of DNA damage in cells.

In

the attempt

to

quantify biologically effective doses

of

ambient

tfV

radiation, efforts have been made to develop dosirnesic systems using biological materials (1). We have been trying to establish

a

simple, versatile and robust dosimeter, using

U1i

sensitive spores of a mutant strain of Bccillus subtilis (2,3). Bacerial spores are dormant forms of life endowed with survivability under exreme conditions such as high and low temperatures, desiccarion and various chemicals and radiations (4). The resisance to Lv-radiation has been studied by genetic and molecular analyses

of8-srrbrilrs mutana producing highly UV-sensitiVe spores due to the dual delecs in nucleotide excision repair and spore-photoproduct l1'ase (4,5).

The

spore dosimeter

is

based

on lhe

extent

of

inactivation of lhe mutant spores spotted and dried on membrane

690

Nobuo Munakata ef a/.

Wavelength (nm)

300

35{t

10, loo 1o'r

lo'!

1on

1o'

1o€

lo' lot lo{

a--a- --e+---a---r---a

++.--H

o--a---t+---a---.-a

r^J I t

280 300 320 340 360

380 [image:3.612.73.316.56.292.2]

\Yavelength (nm)

Figure 1. Spectral characteristics of the spore dosimeter and the filter sheets. The upper panei shows an irradiance spectrum, the inacrivation specmlm of the spore dosimeter and an effectiveness spectrum obtained

by multiplication of the inadiance and action at 0.5 nm intervals (6). A reference erythema action spectrum (11) is also shoun for comparison. In the lower panel. spectral transmissions of BPS07 and BPS04 in various combinations are shown.

filters. The inactivation is strictly exponential under constant fluence rates, and spectral inactivation rate constants ("inactivation action spectrum") have been determined in detail across the whole range of UV radiation (6). Attempts to apply this dosimetry under various conditions have been pursued (7,8), and comparisons with physical dosimeters have been performed (3,6). The dose rate could be

determined under natural sunlight

within

several minutes of exposure. However,

in

order

to

determine cumulative doses in longer time spans, two problems need be circumvented; one is ro reduce the dose and the other is to protect the samples from various environmental assaults including rain and snow. For this purpose, we used blue polyethylene sheets. exhibiting uniform absorption characteristics in the range of UV radiation, to cover the samples. After several anempts of daily and weekly exposures, we initiated continual monthly measurements at various sites in the world. The results of the accumulated data are presented in this report.

MATERIALS AND

METHODS

Preparations and assays of sitore dosimeter samples. Spore dosimeters were prepared and assayed in the laboratory of the first author as described before (2,3,6). In

briel

the spores of Bacillus srbnlrs strain TKJ63I:

(uvrB splB his met leul were mixed wirh 0.67o molten low-melting agarose (SeaPlaque; Cambrex, Rocklend, ME). Four

5 gL

aliquots

of

the suspension, each containing approximately l0o spores. were spotted on a membrane filter (pore size,0.1 prm: A0l0A025A: Toyo Roshi, Tok1o, Japan) in a l2-place filtering device under a weak vacuum. Two spots rvere covered with two pieces of black paper and served as unexposed controls. Each filter was anached with adhesive labels to a glass cover of a slide

mount. The samples were covered and dually wrapped

with

blue pol;"ethylene sheets described below and placed in a plastic slide holder. The samples could be stored in dark either betbre or afier the exposure as

long as needed. They were sent to each exposure site by mail. .{frer exposure, the samples were sent back for lhe assay. Each spot was cut out

and placed in a test tube. One milliliter of water was added to the tube and heated at 75"C for l6 min with vigorous shaking ar the midpoint. An eliquot of the suspension was overlaid on a casamino acid-supplemented minimal agar plate. The colonies were counted atter ovemight incubation at 37"C. The numbers ofcolony formers oltwo exposed spots were divided by those

of

trvo unexposed spots

to

obrain the surviving tmction. The spore inactivation dose (SID) was derived lrom the absolute value of natural logarithm of the surviving fraction: SID = _{-ln(surviving} fraction). \lonthly SID va.lues were calculated by multiplying the factor of rcduction derived for each combination of filter sheets as described below.

Characteri:ation of blue polyethylene sheet. Appropriate filter materials

that also served as protection against adversities were sougjrt by deter-mining optical properties of various materials found in the laboratory and at home.

It

was found that blue polyethylene sheets used as garbage bags exhibited favorable propenies. We ordered two rypes of the sheet (Umeya Sangyo. Tokyo, Japan) with different thickness, one 0.04 mm and another 0.07 mm (referred to as BPS0.I and BPS07, respectively). The ransminance

of BPS04 and BPS07 was determined using the faciliry of Okazaki Lar-se Spectrograph (9) with the sheet fining tightly over rhe photodiode. The results of the transmittance characreristics are shown in Fig. I together with the "effectiveness spectnlm" obtained from solar irradiance and action specra (5,10, I 1). The measurements indicare that in the range of pertinent

UV wavelengths (290-365 nm). the transmiuance was reasonably uniform for each and various combinations of the sheel and that with rwo layers of

BPS07 and BPS04. the transmission could be reduced to less than 0.17o. Exposure condirions and sites. Samples were placed horizonally at a place without shadin-e at any time on the roof of a building at each mea-surement site. Exact coordinates are known from the following sies: Tokyo (l,Iational Cancer Center [35'41'N, 139'16'E] and Rikk-vo University [35"J,*'N, 139'42'E] since April 2002), Marumoto (Shinshu Universiry

[36'15'N, 137"58'E]), Brussels (Belgian Instirute for Space Aeronomy

[50"48'N,

4'21'E]),

Thessaloniki

(Arisode

Universiqv F0'31'N, 22"58'El), Sdo Maninho da Serra near Santa lvlaria (hereafter refened to as Sio Maninho) (Sourhem Space Obsen-amr-v [39"26'5. 53"a9Tf]) and Taipei (National Taiwan Universiry [25'01'N. 121"32'E)\. For other sites, the coordinates were read out from maps. E.rcept for Masumoto (aldtude, 60O m). Sdo Maninho (altitude. 468 m) and Amman (akitude- 760 m). the altirudes of the sites were less rhan 150 m above sea

level-O:one column. The values of monthly average column ozone in Dobson unirs (DLD were obtained from the darabase of Totai Ozone }tapping Spectrometer (TO:VIS) (12). The means

of

lf

months from Jaauary to December were used as the average yearly amouns.

RESULTS

AND

DISCUSSION

Project outlines

After preparative studies of daiiy and weekly exposures.

de

tests

for

monthly exposures were conducred ar Tokyo, Jakana and Denpasar starting in September 1998. These initial tests- using three sets of sample slides with different combinarions of the filter sheets.

indicated that carc should be raken for the sample prepararions to cope with climaric adversities, especiall]' heavy rains. like those encounaered by the frequent squalls in lndonesia and the occasional ryphoons in Japan. Also. the doses at Denpasar were higher than anticipated. Therefore.

it

was decided

that

the samples were wrapped with two layers of the polyethylene sheet (8PS07). using plastic tapes and sealed

*'ith

an electric sealer. Funher filtering was provided with additional layers of BPSOI or BPS07- It $'as also decided that, ar tropical sites. four sample slides for each month should be emplo!eC to cover the hi-eher doses

present-Since January 1999, five sites in Japan and two sites in Europe have been added to initiate continual monthl-v measurements. After this time, several sites including rwo in South America have been included, but the operations at some sites u'ere discondnued due to personnel changes ofaddresses. In this repon- we collecred the data at the sites that provided the data for more than

I

1'ear. The data

could not be collected in some months due to various accidents of both natural causes (samples lost or spoiled in storms. de$rc:'ed by I

1d

to'

€

ron 3

tn'

i

,nt

i

lot ; 10'

?

l0'

-* 10{

g

1(rt

3

1oo

2

E

1o'r

.i,

lo'2

- -a-- BtSlg +851tr --.-.Bmt'&1fi +862ff --a--86ffi+lg +161ff1*

bids,

etc.) and human factors (exchange failures. mail troubles, etc.). Some samples in tropical or subtropical sites could not be

fully recovered due possibly to the combinations of extensive heat and rain. In all, there were 8:1 missing data points out of a total of 980 months. In the estimates of yearly total values, missing points were filled by the means of two values of the same months in the nearest years. Summarized data are shown in Table 1. The data of

the

monthly

values

in a

tabulated

lorm are

available as supplementary material from the website.

Nlonthly and y'early doses at each site

Ettrope. Three European sites. Oulu (65.0"N), Brussels (50.9.N) and Thessaloniki (40.5'N). _{exhibited monrhly doses dependent} primarily on the solar latitude. High values prevailed from June to August, indicating the contribution of column ozone amounts lhat peaked in April and declined to October. Although only

i7

months of data were available at Oulu,

it

exhibited the lowest values

ol

UV doses'in the present work. Because the sun does not rise above 10o during December and January and the outside is covered with thick snow, it was considered not practical to determine the doses,

and we assumed the doses were much lower than

the

19 SID observed in February. Mean yeariy doses with standard deviation in parentheses were 3200 at Oulu.5900 (+1000) ar Brussels and 16 500 (+3669) at Thessaloniki.

Japan. Data from six sites on four islands

of

Japan, covering about

l7o

in

latirude were available.

Ar

Sapporo (43.1"N) on Hokkaido island, monthly changes were similar to those observed at Brussels with a slightly hi_gher values (about 357o). Ar five sires

on

Honshu Island (Matsumoto. 36.3oN; Tokyo, 35.7.N; Kobe, 34.7'N), Kyusyu Istand (Kiyorake, 31.9') and Okinawa Island (Nishihara, 26.4T.i),

in

rhe majoriry

of

years, the values

in

June were lower rhan those

in

lvla.v and July despite higher solar alti-tudes.

A

possible reason is high precipirarion and cloudiness from June to the middle of July due ro the rainy season. yearly total

doses were berween 10 500 (+2000) at Toky,o and 22 500

(-6300)

at _{Nishihara and exhibired large variarions possibly due}

to

_the variations of summer weather.

Two Asian srres. Two low laritude sites, Amman and Taipei, in addition to Japan and Indonesia were included in the course ofthis project.

At

Amman (32.0'i\i-), rhe data were only for slightly over a year, and dusts, probably due ro sandstorms, often dirtied the sample surfaces. _{This could have reduced rhe dose significantly.} The site at Taipei (25.0"19

in

Taiwan

is

siruared berween the southemmosr site in Japan and the sites in Indonesia. and exhibited hi_eh doses with clear seasonal changes. lvlean yearly dose at Taipei was 29900 (+1100).

Indonesia. The four sites

in

Indonesia are

on

three tropical islands and include padan-g _{(0.9"S) on Sumarra, Jakarta (6.2"5) and} Yo_eyakana (7.8'S) _{on the island of Java, and Denpasar} (8.6"5), the

main city of Bali.

All

monthly doses _',vere higher rhan l0OO SID (except one

in

Jakarta) and

did

not exhibir clearly demarcated seasonal changes. In the inirial year,

it

was noticed that the values at Jakarta were consistently lorver than those

of

Denpasar. Uporf the addition of rwo more sites at Yogvakana and padang. it became evident that _{relatively low values were resricted to Jalcana- [n all.} the mean ralio

of

monthlv dose at Jakarta to Denpasar was 0.54 (+0.29 _{SD; 58 data points). The measurement site in Jakana was} in the middle of city with hi_eh

rafiic

and industrial activiries. and ir was likely that the atmospheric transparency was _{much lower} than that at Denpasar or other places due to poliutants in the air. At these

Photochemistry and Photobiotogy,

2006,82

691

sites. the yearly climare typicatly consisred

ol

dry (fiom May to

September) and wet seasons (tiom October ro April). Also,

it

is known that the amount

of

precipitation is erceptionally high at Padang. The lar_ee variations seen in Padang during rhe wet season seemed to depend on the variability in the arnount ol precipitation. The values _{of monthly SID higher than 8000 w.ere observed only in} lndonesia, and these hi_eh values were encountered frequently during August to January. lvlean yearly doses *.ere between 39 400

(-8900)

at Jakarra and 77 800 (+ 15 700) at Denpasar.

South _{America. Measurements at a site in southem Brazil, Sdo} \faninho (29.5"5) and a site ar the sourhem end of South America at Punta Arenas (53.2'5) in Chile were initiared in the yean 2000 and 2001. respectively. Borh sites exhibited seasonal changes opposite to rhe Northern Hemisphere. The site at punta Arenas has

otten been covered by the extension of the .{ntarctic ozone hole during spring monrhs (13,14). Data on some crucial monrhs were rnissin_e, and the extent

of

the effect was nor clear at this point. \lean yearly doses were 5 100 (+700) at punra Arenas and 23 600

(-5000)

at Sao Marrinho.

Ranges of biologically effective solar UV doses

ltonthly

doses of ambient UV radiation were derermined by spore dosimeters ar 17 sites covering a wide range

of

latirude.

At

the exuemes, doses at subarctic and tropical sites diverged from less than 20 units of SID (at Oulu from December to February) ro more than 12000 (at Padang

in

October), represenring a difterence of about 600-fold. This striking variabiliry represenrs a unique learure of solar IJV radiation as _{an environmental factor affecting living} organisms. Seasonal changes

of

the doses are evident at all sites except in the tropics. In rhe tropics, high monthly doses (i.e. >8000 unia of SID) never encountered elsewhere could occur frequently tlroughout a y'ear. Thus, organisms are under a constant .,radiation burden" without long periods of recovery. The ratios of the doses trenveen _{the maximum to minimum month tend to increase at the} sites of _{higher latitudes. Larger ratios indicate the} changes during spring and aurumn months are steeper.

This

_rype

of

seasonal change also should be taken into account in the assessment of bio-logica.l effecs and possible adaptive sreps on pan of rhe organisms e.rperiencing these chan_ees.

There are three major factors affecting the doses

of

solar LfV radiation: the solar altitude. the amount of column ozone and the air opaciry. The ladrude at each sire represents solar alrirudes wirh daily and seasonal periodicity. Daily and monrhly averaged amounrs

of

column ozone at each site within latitudinal (1.01 and lon-sirudinal

(1-5')

grids have been determined by TONIS and are publicly available

(ll).

The most

dilficult

facror

for

quantificarion

is

rhe

tnuxparencv of the air. The amounts and nature of water vapor in air. taliing shape as rain, snow. clouds and humidiry, are considered to t'e imponanl but these could not be represented as simple values. \foreover. there are cases in which air pollution plays a si_enificant

role

in

reducing the doses. Thus, fbllowing comparisons

of

the dos* and analvses of the changes during these vears are pror.isional

rliih

resard to meteorological and climatic aspecs.

The resuls show a decisive globat partem of latirudinal depen-dence. _{The vearly dose exhibited more rhan a}

j0-fbld

difference tenteen the subarctic

city

of

Oulu and the tropicai Denpasar.

-{lthough there could be some places of more e\treme UV

692

Nobuo Munakata et a/.

Table

l.

Summalv of solar UV doses determined with spore dosimetry by location

Oulu

Brussels Thessaloniki

SaPPoro \'Iatsumoto Tokyo Kobe Kivoteke

Site

Coordinates

Measurement Period

Ivlean monthly SID* Januar.v Februar,v NIarch Apnl May June Juty August September October November December

Minimum monthly SID

lvlaximum monthlY SID

Yearly total SIDf 1999

2000 200 1

2002 2003 2001 65.0"N. 25.-i"E 1000/05-200

l/l

l

ND t9 85 219 t1) 71r 1021 831 290 tt7 _r0 ND

1 191 (l0oo/08)

50.9"N.

r999/0

i-)nru/t 1

11 55 r50 367 767

I 04i

I 280 t246 565 ta5 119 IA l6 (20Mlo2) r629 (2003/08) 5327 (0) 6057 (0) 5566 tnr 492't (0) 7779 (2) 59 19 (0)

40.5'N. 2i.0"E r999l0r-2001112 r05 153 !al 129 I 2094 3524 3-109 3089 128 1

i87 255 86 60 (2000/12) 4863 (2004/06)

16 5+3 (2) 164J4 (3) 14689 (1) 12032 (1) 20651 (1) 18400 (1)

.13.1'N. 111.3'E t999l0t-200+105 63 98 187 686 1016 t116 160?

1 i89

708 198

li0

68 ts (2001/Ol) 2371 (2001,08)

7202 (2t 7813 {1, 882-t ( I )

6861 (-r) 8i35 (o)

\D

)O.J r\.

l 18.0"E 1,99911-,onl/l, 175 111 101 892

l 508 l-l0l 26t9 2503 ll12 +l+ )12 184 5J {t999102) 1529 (2001/07)

10-r82 (0)

9i 16 (0) 12589

(l)

8i33 (0) 160.+l (0) I r 615 r0I

35.7"N. r 39.7"E

t999l0t-2O04lt2 156 238 299 850 1336 12.51 2210 2t'76 982 !?1 296 182 101 (1999102) 3\46 (1004/07)

9579 (0) 953:l (0) 896e (0) 9257 (r)

1-r 030 (0)

tr712 (0)

1.1.7"N. 135.2"E t999l0t-20041t2 161 285 361 117.1

1 393

r'730 2'736 2508 rr97 515 296 236 89 (2004/01) 4168 (2003/06)

I I 306 (2)

l r 204 (1) i I 072 (2) 10438

(l)

16 193 (2)

1 3 236 (0)

31.9"N,

li

l..l"E t999l0t-1004112 289 355 14t t 152 t6t9 2229 3400 2622 1320 840 436 218 9'l (2003/12) 4986 (2001/07)

10 26e (0) 12744 (1) 16387 (0)

1-r 028 (2)

2r 379 (3) 17071 (0) ND

N:D

iriR ra) ND ND ND

*ND = Not determined: SID = spore inactivation dose.

iln

parentheses, the number of monrhs with missing data are shown. The estimates *'ere made from the mean of the values in the nearest years'

an exponential dependence on the latitudes

in

both hemispheres

with

almost equal exponent values

of

-0.051

and -0'0-19 for Northern and Southem Hemispheres, respectively' This indicates that the doses double with approximately every 14 degree decrease

in

latitude. The dose at Jakafia. however,

"vas notably lower than

Yogdk rb

Sito Mstitrho

PugArd

\

O"f"\

- -

-y ₌ &{329' e^(-0.05132,{x): R= 0.92011

-:y

= E1530

* _{e^(0.O1946x) R= 0463E2}

1d-.f

-80 -60

-40

-20

2{' 40 60

E0 [image:5.612.74.587.63.376.2]

Latitude (degree)

Figure 2. lvlean yearly doses against latitudes. The bars represent the ranges of the ma.timum end minimum values'

the value obtained from the rcgression, suggesting the effects

of

atmospheric pollution. The generally lower values in Japan seem to be due to the monsoon climate of high humidity and precipitation in summer. The latitudinal gradient seems primarily determined by solar altitude. However. because the amount of column ozone is generally smaller at lower latirudes in both hemispheres' this fact should also be taken inlo account.

Yearly changes of column ozone amounts and

UV

doses

During rhe period of this project' yearly average values of column ozore seemed to have declined steadily at most places as shown in the righr panel of Fig. 3.

in

linear regressions, all the siles except Padang exhibited negative coefficient vaiues. The values (DU change per yeari

r

referring to lhe linear correlation coefficient) smaller than

-2

were obsewed

at

Sapporo

(-3.2;

r :

0'83)' Brussels

(-2.8;

r =

0.88). !{atsumoto

(-2-6;

r :

0'89)'

Seo rv-taninho

(-2.6:

r =

0.74), Tokyo (-2.0;

r:0.75)

and DenPasar

(-1.&

r =

0.7S). On the other hand, all

ol

the sites exhibited an increa.se of yearl;- total doses in either linear or exPonendal regres-sions- The largest yearly incr€ase of SiD in linear regressions was ob=n'ed at Denpasar (5600:

r:

0.67). followed by Padang (3600;

r :

0.+l).

Five

sites

in

Japan exhibited significant increases: Nishihara (2100:

r:

0.6i), Ki1-otake (1600:

r

= 0.80). Kobe (650:

r:0-56).

Matsumoto (750

r=

0.5?) and Tokyo (710:

r:0.67)'

Small:r and less signi{icant increases in UV doses werc observed at SapFrro (170; r = 0.32) and Brussels (210; r = 0.40)' where the decre:-ses in ozone arnounts \rere most significant. The global trend 10"

U)

[image:5.612.40.325.411.705.2]

Photochemistry and Photobiology, 2006,

82

693

Table

l.

E-ttended

Nishihara Taipei Padang Jakarta Yogyakarta Denpasar Sdo

Maninho

Punta Arenas

26.4"N,

r l7 00F

t999/0t-20041t2

429 540 742

t'7 52

)))1 3118 4504 3919 2277 r439 795 416 164

( i99910

l)

59'16 (2o0u07)

16 676 (0) 16 155 (0) 24 065 (0) 23205 (t) 33525 (r) 2t227 (0)

32.0"N, 35.9"E 2001106-7003102 401 61i 663 908 1703 3986 2775 1554 )1 J1 1066 561 ND 401 (2002/01) 4974 (2001/06) ND ND ND l6 378

(l)

ND ND 25.0'N. l2 r.5'E 200v11-2M4112 351 JJ6 719 I 188 3987 6126 6638 5605 2216 1406 855 143 154 (20c,rlot) 7307 (2002/M) ND ND ND 30666 (1) 30376

(l)

28 589 (0)

0.9's, 100.4"E t999/tr-2M4lt2 6866 5566 566 r

5491 4240 4503 6325 8942 7087 88 17 5250 6116 1116 (2002/05) 12255 (2001/09) ND 55 847 (1) 87 856 (2) 64007 (2) 83439

(l)

75e39 (2)

6.2'S, 106.8'E t999l0t-20o4lt2 3391 2I7 t 2266 4064 2553 1q1 i

28ri 2668 3493 4150 4382 4056

7ll

(teeg/02) 877 t (200r/10)

i4519 iit

32 136 (0) 53 010

(l)

32026 (2\ 47810

(l)

3679r (0)

7.8'S, I 10..1"E 2000/0i-2001/10 1993 5232 7733 62U 490'7 3108 2048 .1306 7734 53i5 3830 ND 1925 (2000/07) 89.+6 (2001/03) i\D

56.11-l (1) ND |aD

i\D ND

8.6'S. 1 15.2'E

1999l0t-?\Mlt2 80i6 5288 6288 6212 4954 5505 5330 70'77 7498 7934 '1672 57 14 227 1

(200210s) 1l 800 (2003/08)

58 628

(l)

63 489 (1) 96047 (t) 't0't37 (r) 92590

(l)

85 522 (3)

29.5'S, 53.8"W 2000/03-20011t2 4.t03 28,{0 2283 t222 623 375 526 807 t265 2155 27 13 J /JU

t25 (2002106) 6s26 (2003/01) ND ND 25 424 (t)

16 948 (0) 28 955 (0) 23234 (r)

53.2"S, 71.0"w 2001/05-20M/t2

7,18

1A I 331 l15 A1 45 66 89 3t2 551 908 l03l 25 (2002/06) 1228 (2003102) ND ND ND 4380 (r)

s832 (2)

5 150 (2)

of ozone decrease and UV increase clearly continued between the years 1999 and 2004.

The pattems of year-to-year variations of column ozone amounts and

IIV

doses did not always exhibit clear correlations at most sites of high and middle latirudes. For example, prominent peaks of {.tV

doses observed in rhe years 200

I

and 2003 at several sites in Japan

did not correspond to column ozone changes. This indicates that the changes of

W

doses in these years were caused primarily by climatic variations. On the other hand, at three Indonesian sites,

two prominent peaks

of UV

doses

in

the years 2001 and 2003 corresponded to the troughs ofcolumn ozone amounts. There seem

to be several reasons why the ne-gative correlations between UV doses and column ozone zunounts are shown clearly in the tropics. Because ozone absorption increases steeply at the lowest wave-length end of the eff'ectiveness spectrum, the loss could manifest itself more severely when the initial amounts were lower as in the Asian tropics. Also, the small seasonal changes

ol

both ozone amounts and

W

doses could contribute

to

the correlation. In

Average Column Oune (DL)

D U $ a I o a +TlEsroniki +llalanorC

-ts 1::l:i.:ra

+geoMalihrc

[image:6.612.58.564.40.380.2]

Year

Figure 3. Yearly toral doses (SID) ar l7 sires (left panel) and average column ozone amounr (DU) at markers without lines ue from the sites with data from only

I

year.

Year

13 sires (right panel). In the lelt panel, rhe solitar,v Yearly Spore Inactivation Dose

[image:6.612.63.571.493.715.2]

694

Nobuo Munakala et a!.

contrast. at other sites, the column ozone amounts exhibit large seasonal changes w'ith a peak

in

spring and a trough

in

autumn. Since yearly total values are largely

lrom

summer months. the changes mi,eht not affect the UV doses directly.

Use of spore dosimetr,v

for

global monitoring of solar UV radiation

ln this work. the biologically effective doses of solar UV radiation were determined with the use

of

spore dosimetry as cumulative monthly values at various sites. One serious problem in giobal UV dosimetry has been that a majoriry of the monitoring stations are situated at high and middle latitudes and measuremenr in tropical environments have rarely been performed despite the expected high doses due to high solar altitude and low ozone amounts. To the best og ouv fuiowledge, this

is

the first result

of

monitoring performed

in

the Asian tropics

in

parallel

to

middle and high latirude places. The monitoring

of UV

doses and column ozone amounts in the tropics is critically important with regard to human health, productiviry and conservation because of the large human populations

in

countries

of

this region and because of the biodi-versity and range of ecosystems encountered there.

Several advantages

of

a

purely biological approach

in

Uv: monitoring have been demonstrated

in

this work. Ivleasurements can be performed an1'where without special facilities. The dose

could be represented by a simple biologically relevant value

di-rectiy

obtained

from

experimenul measurements'

ln

conrrast physical dosimeters require dedicated instruments and resources. and. to derive biological doses, calculations dependent on cenain models. On the other hand, some problems were recognized in order

to

advance long+erm applications

of

the spore dosimeq''-One is the lack of a reliable reference to check the consistency of the inactivation kinetics in the field. We assume the inactivation is strictiy exponential to the accumulated dose, but

it

is difficult to validate the kinetics under the conditions of incessant changes of specral irradiance during a period of a month. The use of the filter sheet was necessary, but

it

introduced another complication. be-cause any change and variability

in

the optical characteristics of the filter sheet could cause large changes in the doses'

pafiicularl)-in

the cases

of

very high doses.

It

is desirable to carry out daiiy monitoring in addition to the monthly one at some sites and to use different types

of

filten

(for example, neutral densiry filters) for comparative purposes. Comparisons with data obtained by physical dosimeters at several sites involved in this project are in progress and

will

be presented elsewhere.

Acknowledgemenrs-This work has been supponed

in

pan by grans-in-aid from Ministry of Educadon. Science, Spons and Culure, Japan

(07303043 and 99042004) and Japanese Committee ior Sun Protection (JCSP). We thank encouragements and helptul discussions by Dr- Keiichi Nozu (JCSP), Dr. George Ront6 (Hungarian Academy' ol Sciences, Buda-pest) and Dr. Cerda Homeck (DLR, Koln).

REFERENCES

1. Renberg, P., G. Homeck, C. Baumstark-Khan and G' Amanatidis, eds' (1999)'Biological UV dosimetry, a tool for assessing the impacr of LIV radiation on health and ecosystems. Air Pollution Research Repon 71. European Commission, Luxembourg.

2. Munakata, N. (1981) Killing and mutagenic action of sunlight upo-n Bacillus subtilrs spores: a doiimetric system..!Iarcr. Res 82.263-268' 3. Munakata. N., F- Morohoshi.

K.

Hieda,

K.

Suzuki. Y- Furusawa.

H. Shimura and T. Ito (1996) Experimental conespondence between spore dosimetry and spectral photometry of solar ultraviolet radiation'

P horoche m. P hotobiol. 63, 1 4-7 8.

4. Nicholson, W.

L., N.

Munakata, G. Homeck. H. J. \lelosh and

P.

Setlow (2000) Resistance

of

Bacillus endospores

Io

exreme terrestrial and extraterrestrial environments. Microbiol. Ilolec. Biol-Rev. 61,548-572.

5. Munakata, N. and C. S. Rupen (1974) Dark repair of DNA conraining "spore photoproduct"

n

Bacillus subtilis. _'Volec- Gen- Genet- 130' 239-250.

6. Munakata, N., S. Kazadzis, A. F. Bais, K. Hieda- G. Ront6- P. Reaberg and G. Homeck (2000) Comparisons of spore dosimeq and spectral photometry of solar-tlV radiarion at four sites in Japan a*1 Europe'

P hotochem. P hotobiol. 72,'739--1 45.

7. Munakata, N., M. Ono and S. Watanabe (1998) lvlonito@ of

solar-UV exposure among schoolchildren in five Japanese cities usiag spore dosimJter and UV-coloring label. Jpn. J. Cancer Res- 89.

I)-245'

8. Munakara, N., K. Makita, D. Bolse6, D. Gilloay and G- Homeck (2000) Spore dosimetry of solar UV radiarioo: applicatioas

o

moni-toring of daily irradiance and personal exposrue- Adl.- Space Res' 26, 199)-2003.

9. Waranabe, M.. M. Furuya, Y. Miyoshi, Y. I-noue.

L

lo:]rashi and K. Masumoto (1982) Design and performance of the Okazaki large spectrograph for photobiological srudies. Photochem- Photobiol. 36-

491-{98-10. Serlow, R. B. (197.1) The wavelengh in sunlight effective ia producing skin cancer: a theoretical analysis. Proc. Narl- Acad- Sci- LI 5 A 71. 3363-3366.

11. N{cKinley, A. F. and B. L. Diffey (1987) A ret'erence

rticn

gectrum for ultraviolet induced erythema in human skin. CIE Res- _'\bre 6-l7-?2.

12. Total Ozone Mapping Spectrometer Ozone Processing Te'n ar \'c'S'di GSFC. Available

at:

http://toms.gsfc-nasa-gov/pub, eptoms/dara-Accessed on

I

October 2005.

i3.

Rousseaux, M. C.. C. L. Ballare. C. V. Giordano' A-

L

ScoF€L

.{ \L

Zima. M. Szwarcberg-Bracchiaa P. S. Searles- M- lv{- Cai.hell and

S-B. Diaz (1999) Ozone depletion and UVB r:diadon: imgact on planr

DN.{ damage in southem South America- Proc- iuatl- Acei Srr" U S A 96. i5310-15315.