Shading Screens for Frost Protection

AGRICULTURAL
AND
FOREST
METEOROLOGY
E L S E VI E R
Agricultural and Forest Meteorology 81 (1996) 273-286
S h a d i n g s c r e e ns f or f ros t pr ot e c t i on l
M. Te i t e l a,*, U. M. Pe i p e r a , y . Zv i e l i h
a Agricultural Engineering Institute, Agricultural Research Organization, Volcani Center, PO Box 6, Bet
Dagan, 50250, Israel
b Ministry of Agriculture, Extension Service, Arava Region, Sapir Center, 86825, Israel
Received 14 February 1995; revised 18 September 1995
Abstract
Shadi ng screens stretched hori zont al l y above the ground, were found ef f ect i ve in reduci ng the
risk of frost damage. The screens reduce the net amount of l ong- wave radiation from the ground to
the sky during the night and thus keep the t emperat ure of the plants under the screens at a hi gher
t emperat ure than ambient. A model for cal cul at i ng the reduction in l ong- wave radiation exchange
bet ween the ground and the sky, due to the presence of a screen, was devi sed and veri fi ed by
experi ment s. The model suggests that three parameters affect net radiation under the screen,
shading percent age of the screen, radi omet ri c properties of the screen and the ratio bet ween screen
area and the ground area beneath it. Of several types of screens that were tested, an al umi ni zed
screen was found to be the most ef f ect i ve in reduci ng frost damage. A si mpl e model for
cal cul at i ng l eaf t emperat ure is offered and used for cal cul at i ng the temperature of an upper leaf.
The experi ment al data and cal cul at i ons show that during the night the t emperat ure of the l eaves is
l ower than the air t emperat ure and t herefore frost prot ect i ve devi ces should be cont rol l ed
accordi ng to l eaf t emperat ure and not air temperature.
1. I nt r o duc t i o n
Th e gr e a t e s t a gr i c ul t ur a l r i s k i n c o n n e c t i o n wi t h l o w t e mpe r a t ur e s is f r ost , wh i c h c a n
c a us e s e v e r e de s t r uc t i on o f f r ui t , v e g e t a b l e s and pl ant s. Th e s e ns i t i vi t y o f a c r op t o l o w
t e mpe r a t ur e s d e p e n d s on ma n y f act or s , i n c l u d i n g t he s e ve r i t y o f t he t e mp e r a t u r e dr op
a nd f or h o w l ong t he c o l d per s i s t s . Pl ant s pe c i e s di f f e r gr e a t l y i n t he i r s us c e pt i bi l i t y t o
c hi l l i ng i nj ur y. T wo ki nds o f f r os t s ma y be di s t i ngui s he d: r a di a t i on f r os t a nd a d v e c t i o n
f r ost . Th e f o r me r oc c ur s on c l e a r ni ght s wh e n a l ar ge a mo u n t o f he a t i s r a di a t e d t owa r ds
* Corresponding author.
I Contribution No. 1546-E, 1995 series, from the Agricultural Research Organization, The Volcani Center,
Bet Dagan, Israel.
0168-1923/96/$15.00 Copyright © 1996 Elsevier Science B.V. All rights reserved.
SSDI 0168- 1923( 95) 02321- 6
274 M. Teitel et al . / Agricultural and Forest Meteorology 81 (1996) 273-286
the sky, and its occur r ence is general l y spot t y; the later results f r om the i ncursi on of col d
air masses. The damage due to radi at i on frost differs f r om that due to advect i on frost
mai nl y in its degree. Plants that are killed by advect i on frost are usually onl y partially
damaged by radiation frost (Critchfield, 1966).
Pr event i on of crop damage due to radi at i on frost is mor e feasible than advect i on
frost. Dur i ng radi at i on frost, onl y a thin l ayer of air i mmedi at el y above the gr ound is
cool ed whi l e the over l ayi ng layers are war mer ( Rosenber g et al., 1983; Oke, 1987; Gat
and Karni, 1993). A light wi nd is general l y suffi ci ent to mi x the cool er and war mer
l ayers and thus dispel the frost. Prevent i on of radi at i on frost can be achi eved by
br eaki ng up the i nversi on that accompani es intense ni ght -t i me radiation. This is gener-
ally accompl i shed by stirring the air, heat i ng it, pr ovi di ng a prot ect i ve bl anket of smoke
or by any combi nat i on of these. One of the ol dest proposal s for frost prot ect i on ( Br ooks,
1961) is " . . . artificial cl oud screeni ng to r educe the rate of heat loss to the col d s ky. "
The pur pose of the current research was to st udy the feasibility of usi ng convent i onal
shadi ng screens, st ret ched over the crop, to r educe l ong- wave radiation l oss to the sky
duri ng the night, and thus reduce the ri sk of frost damage.
2. Experimental setup
The exper i ment s were carri ed out in the Ar ava regi on (30. 25°N, 35. 15°E, alt. 90 m)
in the sout her n part of Israel in t wo successi ve wi nt er seasons (i.e., wi nt er 1992/ 1993
and wi nt er 1993/ 1994) . A commer ci al area of about 0.25 ha of capsi cum was di vi ded
into 12 adj acent regi ons each of an area of 10 X 20 m, over whi ch di fferent t ypes of
screens were st ret ched at a hei ght of about 2.5 m above the ground. Two screens f r om
each t ype were tested for repetition. The screens are identified by a commer ci al
desi gnat i on t hat states the shadi ng per cent age of the screen and its col or (e.g., ' 30%
bl ack' ) . They are manuf act ur ed f r om pol yet hyl ene sheets that are pr oduced by a
bl ow- bubbl e ext rusi on process. Strips are cut f r om the sheet and stitched to f or m a
woven screen. Al umi ni zed screens were al so tested; t hey are pr oduced by coat i ng a
pol yet hyl ene film wi t h a thin l ayer (0.2 ~m) of al umi num. Strips are t hen cut f r om the
al umi ni zed fi l m and stitched to f or m a screen.
Ai r t emperat ure, T a (°C), gr ound t emperat ure, Tg (°C), l eaf t emperat ure, Tj (°C), and
net radiation, q, ( Wm - 2 ) were measur ed si mul t aneousl y under each screen. One sensor
was used for each t ype of measurement . The sensors were pl aced at the cent er regi on
under each screen to el i mi nat e i nt erference f r om adj acent screens. A schemat i c represen-
t at i on of the experi ment al setup is shown in Fig. l(a). The tested screens were:
1. ' 20% whi t e' ,
2. ' 30% bl ack' ,
3. ' 40% bl ack' ,
4. ' 50% bl ack' , and
5. ' 50% al umi ni zed' .
A phot o of t wo screens, ' 50% bl ack' and ' 50% al umi ni zed' , is shown in Fig. l(b).
Ambi ent condi t i ons (wi nd and t emperat ure) were moni t or ed cont i nuousl y. Gr ound
t emperat ure and l eaf t emperat ure were measur ed onl y in the wi nt er of 1993/ 1994.
M. Teitel et al. / Agricultural and Forest Meteorology 81 (1996) 273-286
( a )
Tam ---r
500
0 up anemometer
~ Net radiometer
qr
0
o Thermocouple
Tl
5 0
( b )
2500
275
Fig. I. (a) Schematic view of experimental setup. %, net radiation; qr, reference net radiation; ql and q2,
thermal radiation flux from the sky and the ground, respectively; T a, air temperature; Tj, leaf temperature; Tam,
ambient temperature; u, air velocity. Dimensions are in mm. (b) Photo of ' 50% black' screen (right) and ' 50%
aluminized' screen (left).
S i n c e l o w t u n n e l s ar e a l s o u s e d t o g r o w c a p s i c u m i n t he Ar a v a ( i n t hi s c a s e t he
s c r e e n s a r e s p r e a d o v e r t h e c r o p s ) t e mp e r a t u r e wa s a l s o me a s u r e d at c r o p l e v e l i n s i d e a
l o w t u n n e l , ma d e o f ' 3 0 % b l a c k ' s c r e e n ma t e r i a l f or c o mp a r i s o n wi t h a h o r i z o n t a l
s c r e e n.
2.1. Temperat ure measurement s
T h e a i r t e mp e r a t u r e u n d e r e a c h o f t he t e s t e d s c r e e n s wa s me a s u r e d b y me a n s o f
c o p p e r - c o n s t a n t a n t h e r mo c o u p l e s h a v i n g a wi r e d i a me t e r o f 0. 5 mm. T h e t h e r mo c o u -
276 M. Teitel et a l . / Agri cul t ural and Forest Met eorol ogy 81 (1996) 273- 286
ples were pl aced within the crop at a hei ght of about 0. 4 m above the ground. Special
caut i on was gi ven to place the t her mocoupl es among the leaves to prot ect t hem f r om
radiation. In addition, the t emperat ures of upper l eaves of the crop were measur ed under
part of the screens, by inserting fi ne-wi re t her mocoupl es (wire di amet er of 0.1 mm) into
the mi dri bs of the leaves. Gr ound t emperat ure under the screens was measur ed by
i nsert i ng a t her mocoupl e into the gr ound to a dept h of about 5 cm. All t her mocoupl es
were sampl ed once per mi nut e by a Campbel l Mi cr o- l ogger . The col l ect ed dat a were
aver aged and stored in memor y, once ever y 30 min.
2.2. Radi at i on me a s u r e me n t s
Net r adi omet er s were used to measur e the net radi at i on under the screens duri ng the
night. The radi at i on was measur ed by means of Radi at i on Bal ance 8110 net radi omet ers
( 0 . 3 - 6 0 I, zm) manuf act ur ed by Phi l i pp Schenk. The sensitivity of the pr obes is about
0. 009 mV W- ~. The radi omet ers were posi t i oned at a hei ght of 0.9 m above the gr ound
(about 0. 2 m above the plants) and cl ose to the t her mocoupl es that measur ed air
t emperat ure under each tested screen. Net radi at i on wi t hout a screen (for reference), qr,
was meas ur ed by cut t i ng a hol e in one of the screens and pl aci ng the r adi omet er at
screen level. Ambi ent t emperat ure, T~m, and wi nd vel oci t y, u, were measur ed by a
radi at i on-shi el ded t her mocoupl e and cup anemomet er , respect i vel y, at a hei ght of about
0.5 m above the screens.
3. Theoreti cal consi derati on
Cr ops are usual l y gr own in the field under t wo mai n conf i gur at i ons of shadi ng
screens: (a) a flat hori zont al shadi ng screen; and (b) a hemi cyl i ndri cal tunnel made of
shadi ng- scr een material. The f or mer is mor e popul ar among gr ower s in Israel.
Unl i ke the case in whi ch cont i nuous films t hat are st ret ched above the crops, not all
the upwar dl y di rect ed l ong- wave radi at i on l eavi ng t he crop and gr ound is i nci dent on the
shadi ng screen, because of the gaps in the screen. Some fract i on of the radi at i on passes
t hr ough the gaps t owards the sky. The amount of radi at i on that passes t hr ough the gaps
is dependent on the shadi ng per cent age of the screen, 05. It is defi ned here as
05 = A s / A t , where, A s (m 2) is the bl ocked area of t he screen (proj ect ed area of the
fabric, by normal proj ect i on) and A¢ (m 2) is the total area of the screen (i ncl udi ng the
gaps). In addition, we defi ne a radi at i on shape factor, since radi at i on heat t ransfer
bet ween t wo surfaces is dependent on the shape f act or bet ween t hem ( Love, 1968). The
radi at i on shape factor, r/j k is the fract i on of di ffusel y distributed radi at i on l eavi ng a
surface Aj (m 2) and r eachi ng surface A k (m 2) di rect l y. The first subscri pt of the shape
f act or denot es the emi t t i ng surface, while the second subscri pt denot es the surface
r ecei vi ng the radiation.
The total net radi at i ve energy, Q, ( W) t ransferred to a surface by t hermal radi at i on is
defi ned as the di fference bet ween the i nci dent flux densi t y and the emi t t ed fl ux densi t y,
i.e., Q, =A( qi - qe), where, A (m z) is the area of the surface, and qi (Wm-2) and qe
( Wm - 2 ) are the i nci dent and emi t t ed fl ux densities.
M. Teitel et al. / Agri cul t ural and Forest Met eorol ogy 81 (1996) 273- 286 277
It can be shown ( Tei t el and Segal , 1995) t hat t he net r adi at i on under a scr een f or mi ng
a t unnel is gi ven by:
1 - 4 ' + / 3 4 '
( q , - q 2 ) ( 1 )
q" = 1 - (1 - / 3 ) ( 1 - r/$,) 4'
wher e q, ( Wm - 2) is t he net l ong- wa ve r adi at i on; /3 = 0. 5e s + ~-s, is t he t r ansf er f act or
of t he scr een mat er i al whi ch is de pe nde nt on t he l ong- wa ve emi s s i vi t y, es, and
l ong- wa ve t r ans mi s s i vi t y, ~-s, of t he scr een mat er i al ; r/~f is t he r adi at i on shape f act or
bet ween t he scr een and t he gr ound and i s dependent on t he r at i o bet ween scr een ar ea A¢
and t he gr ound ar ea beneat h it; q~ and q2 are t he l ong- wa ve f l ux densi t i es f r om t he sky
and t he gr ound, r es pect i vel y. Fr om Eq. (1) it f ol l ows t hat net r adi at i on under a scr een is
de pe nde nt on t hr ee par amet er s . Na me l y, s cr eens ' s hadi ng per cent age, r adi omet r i c
pr oper t i es of t he scr een and t he r adi at i on shape f act or whi ch is der i ved f r om t he r at i o
bet ween t he t ot al ar ea of t he scr een and t he gr ound ar ea beneat h it. Not e t hat wi t hout a
scr een t he net r adi at i on is q , = q~ - q 2 .
The f ol l owi ng as s umpt i ons wer e ma de in t he der i vat i on of t he expr es s i on for net
r adi at i on under scr eens:
1. al l sur f aces are i s ot her mal ;
2. al l s ur f aces are di f f us i ve;
3. ai r does not par t i ci pat e in t he r adi at i ve t her mal exchange;
4. t he r adi ant f l ux dens i t i es are uni f or m;
5. bot h si des of t he scr een have i dent i cal pr oper t i es;
6. in t he case of a hor i zont al scr een st r et ched a bove t he gr ound, t he scr een is i nf i ni t el y
l ar ge and is pl aced cl os e to t he gr ound so t hat L >> d, wher e L ( m) is a char act er i s t i c
di me ns i on of t he scr een and d ( m) is its di s t ance f r om t he gr ound; and
7. t he scr een is made of unt wi s t ed st r i ps whi ch are cut f r om a ver y t hi n fi l m.
For t he cas e of a ver y l ar ge, fl at , hor i zont al scr een, "q~f = 1 and t he expr es s i on f or net
r adi at i on r educes to:
q, = (1 - 4' + / 3 4 ' ) ( q , - q2) ( 2 )
Eq. ( 2) can al so be used to obt ai n net r adi at i on under a hor i zont al t her mal scr een
ma de of a cont i nuous f i l m, by set t i ng 4' = 1. The net r adi at i on i s t hen q, = / 3 ( q ! - q2 ).
Thi s expr es s i on is i dent i cal t o t hat gi ven by Bai l ey ( 1981) and Ams e n (1975).
A t r ansf er f act or F (0 < F < 1) can next be def i ned as:
q,
F = - - ( 3 )
ql - q2
Fr om Eqs. (2) and (3) we get for a hor i zont al scr een:
r --- 1 - 4' +/ 3 4 ' ( 4 )
The t r ansf er f act or r epr es ent s t he f r act i on by whi ch a scr een r educes t he t her mal
r adi at i on ener gy exchange bet ween t he gr ound and t he sky. As expect ed, Eq. ( 4)
e mpha s i z e s t he des i r abi l i t y of a scr een wi t h l ow emi s s i vi t y, l ow t r ans mi s s i vi t y and a
hi gh val ue of s hadi ng per cent age when l ow r adi at i ve heat l osses ar e r equi r ed.
278 M. Tei t el et al. / A g r i c u l t u r a l a n d For e s t Me t e or ol ogy 81 ( 1996) 2 7 3 - 2 8 6
4 . E x p e r i m e n t a l v e r i f i c a t i o n
Experiments were carried out in order to verify the theoretical results. Two types of
woven screen were used, made of al umi ni zed polyethylene and of black polyethylene.
Commercial as well as i n-house-made screens were tested. The in-house screens were
made by punchi ng holes in a polyethylene continuous film. Screens with various
solidities were tested. It should be emphasized that the error in evaluating the shading
percentage of the commercial screens is relatively large. In the present experiments, the
shading percentage was det ermi ned by stretching the screens over an openi ng in a black
box, and i l l umi nat i ng them from above. Light intensities beneath and above the screen
were measured with a luxmeter and the ratio between these measured intensities was
equal to I - 4). The commercial screens, each measuring 3 × 3 m, were stretched at a
height of about 1.5 m above the ground and the in-house screens, each measuring
0.6 × 0.6 m, were stretched at a height of about 15 cm above the ground. The net
radiation under the screens was measured by means of net-radiation radiometers. The
radiometers were placed 30 cm below the commercial screens and about 5 cm below the
in-house screens, to mi ni mi ze their exposure to radiation from surroundi ng objects. An
additional net radiometer was positioned beside the screens to measure ambi ent net
radiation, for reference. The data from the radiometers were sampled by a 21X
Campbell Micrologger every mi nut e and averaged over 5-mi n intervals. The averages
were stored in memory for further data processing on a personal computer. Data
collected on 15 nights (between 18:00 h and 05:00 h) were used to calculate the
averages and the standard deviations of the net radiation. Fig. 2 shows the transfer factor
obtained from the experimental data. The error bars on the experimental data represent
the standard deviation. The theoretical values, calculated from Eq. (4), are also presented
° . 8 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
0
¢.9
' ~ 0 . 4
a~
0 . 2
0
o o 1 1 o 1 2 o 1 3 o 1 5 o 1 6 o 1 7 o l s 1
Screen s h a d i n g percentage, ¢
Fig. 2. Transfer factor, F, of a horizontal screen. (a) black screen, Eq. (4) /3 = 0.55; (b) aluminized screcn, Eq.
(4) /3 = 0.12; ([]) aluminized screen, experimental; ( × ) black screen, experimental.
M. Teitel et al. / Agricultural and Forest Meteorology 81 (1996) 273-286 279
o
, , . a
B
6
4
2
o
- 1 0
1 B : O 0
v ' - . ~ 1 6 / 1 / 9 3 t""
-~ '~ ,~'
l ' ~,~;
J J i i J i i i i i i ; i i i i i i t i i t
O 0 : O O 0 6 : 0 0 1 2 : 0 0 1B:00
T i me ( h)
; 1 7 / 1 / 9 3
e u n e e t 1 7 : 0 2 ( f ) e u n r i e ~ 0 6 : 3 9
0 0 : 0 0 0 6 : 0 0 1 2 : 0 0
Fig. 3. Air temperature, T a, under the tested screens. (a) 50% aluminized; (b) 30% black; (c) 40% black; (d)
20% white; (e) low tunnel; (f) ambient.
in t he f i gur e for compar i s on. The agr eement bet ween exper i ment al and t heor et i cal
val ues is good, cons i der i ng t he uncer t ai nt y i nvol ve d in det er mi ni ng t he s cr eens ' shadi ng
per cent age and al so t hat t he t heor et i cal r esul t s wer e obt ai ned f or t he case of a ver y l ar ge
scr een ( L >> d) . The agr eement be t we e n exper i ment al r esul t s and t heor et i cal r esul t s
appear s t o be a l i t t l e bet t er f or t he a l umi ni z e d t han f or t he bl ack scr een.
Owi ng to t he woven st r uct ur e of t he scr eens, t he r adi omet r i c pr oper t i es of t he scr eens
( es peci al l y at l ow shadi ng per cent ages ) appear t o di f f er f r om t hose of t he f oi l f r om
whi ch t hey wer e manuf act ur ed. In addi t i on, it s houl d be not ed t hat t he t heor et i cal r esul t s
wer e obt ai ned for a s t eady- s t at e cas e wher eas in t he pr esent meas ur ement s t he t emper a-
t ur e of t he gr ound decr eas ed gr adual l y dur i ng t he ni ght . Thes e di f f er ences ma y al so have
cont r i but ed to t he obs e r ve d di f f er ences bet ween t he t heor et i cal and e xpe r i me nt a l resul t s.
The t r ends obs e r ve d in t he t heor et i cal and e xpe r i me nt a l r esul t s are, however , si mi l ar .
The di f f er ence bet ween t he bl ack and a l umi ni z e d scr eens is smal l when t he s hadi ng
per cent age of t he scr eens is s mal l and it i ncr eas es wi t h t he i ncr eas e in shadi ng
per cent age.
5 . R e s u l t s
Ai r t emper at ur es under t he var i ous scr eens, over t he t wo s ucces s i ve col des t ni ght s of
t he 1 9 9 2 / 1 9 9 3 wi nt er season, ar e s hown in Fi g. 3. The f i gur e cl ear l y s hows t hat t he
a mbi e nt t emper at ur e dr oppe d be l ow zer o f or sever al hour s dur i ng t hese ni ght s. It shoul d
280 M. Teitel et al. /Agricultural and Forest Meteorology 81 (1996) 273-286
1B0
1 6 0
. . . . . . - . . . .
Ca) . . . . . .
. . . . . . . f . y ' " "
.
140 ..-
o 1 2 o . (b) / , - ' " " , . / . ;
" ~ ~oo L ' " ' " ( d ) / " ~
o .... j !
s 0 . y ( e ) ( )
,,~ o 6 o ........................... / . . / ~ , ( f ) .
E~ 4020.. " ' " " ' ' ' ......... ~ . . . . , , , , , , , . * * ' * * * " / ' J ' " / ° ' -
o '4 - ' a ' 2 ' 1 ' '
- - - ; ~ ~. 3 4
Te mpe r a t ur e (°e)
Fig. 4. Total number of hours that the air under the screens was below a certain temperature. (a) ambient: (b)
low tunnel; (c) 20% white; (d) 30% black; (e) 40% black; (f) 50% aluminized.
be emphasized that during that winter, subzero temperatures were registered over eight
successive nights. Such a long period of frost is most likely to cause severe damage to
crops. Indeed, severe damage was observed under all type of screens except under the
al umi ni zed screen. The lowest temperature, about - 8°C, was registered on 16/ 1/ 93, as
shown in Fig. 3. Large fluctuations of temperature were observed on that night. These
fluctuations are apparently due to the stirring of air caused by local winds (helicopters
were also operated duri ng that night to help to stir the air).
The air temperature under the al umi ni zed screen was consistently higher than that
measured under the other screens, over the entire experimental period. It should be
noticed, however, that the differences in air temperature among all other types of the
screens were negligible. It appears that the temperature under the low tunnel was lower
than that under all other screens and for much of the time was about 2°C lower than that
measured under a ' 20% white' screen.
The total number of hours, over a period of twelve nights, that the air temperature
under each screen was below a certain temperature, was calculated and is presented in
Fig. 4. These results suggest that the temperature never decreased below zero under the
aluminized screen, while the maxi mum number of hours - - about 96 - - below zero,
were registered under ambi ent conditions. For the other screens, i.e., ' 20% white' , ' 30%
black' , ' 40% black' and low tunnel, the total number of hours below zero were 24, 19,
18 and 54, respectively. It therefore appears that low tunnels are least effective in
M. Teitel et a l . / Agri cul t ural and Forest Meteorology 81 (1996) 2 73-286 281
1 0 1 17,,,93 f
- 20 -
(~)
"~ - 4 0 -
,... t u . . . . . . . _ j
/ ' ~/
J
f (e) /
0J
i . z
- 60
sunset 17:02 sunrise 06:39
I oB~o
~ 7 0 i i i I i i I ~ i I I i i i I I I i i 4
12:00 1B:O0 00:00 : 12:00
T i me ( h)
Fig. 5, Net radiation, qn, under the screens. (a) 50% aluminized; (b) 40% black; (c) 30% black; (d) 20% white.
pr event i on of f r ost damage. The f i gur e al so s hows t hat under al l s cr eens except of t he
a l umi ni z e d one, t he l owes t t emper at ur e was - 3 ° C . The t emper at ur es under t he scr eens
di d not decr eas e f ur t her even at a mbi e nt t emper at ur es l ower t han - 4 ° C . The f i gur e al so
s hows t hat t he ef f ect i venes s of t he s cr eens i ncr eas es as ambi ent t emper at ur e decr eas es ,
i. e. t he r at i o bet ween t he number o f hour s be l ow a cer t ai n t emper at ur e wi t h a scr een t o
t hat wi t hout a scr een, is decr eas i ng as a mbi e nt t emper at ur e decr eas es . It can t her ef or e be
i nf er r ed t hat t he ef f ect i venes s of t he s cr eens is non- l i near l y de pe nde nt on a mbi e nt
condi t i ons .
Net r adi at i on under t he var i ous scr eens i s shown in Fi g. 5. Whi l e t he net r adi at i on is
pos i t i ve dur i ng t he day it be c ome s negat i ve dur i ng t he ni ght . The negat i ve val ues wer e
e xpe c t e d si nce at ni ght t he gr ound t emper at ur e is hi gher t han t hat o f t he sky and
t her ef or e heat is r adi at ed t owar ds t he sky. The aver age r adi at i on me a s ur e d by t he
r ef er ence r a di ome t e r (t he one pl a c e d in t he hol e) , qr, was about - 7 4 Wm - 2 dur i ng t he
ni ght . Ne ga t i ve val ues wer e al so me a s ur e d under t he scr eens, but t hei r abs ol ut e val ues
wer e l ower t han t hose me a s ur e d by t he r ef er ence r adi omet er , as expect ed. The dat a
s ugges t t hat net r adi at i on under t he a l umi ni z e d scr een has t he l owes t abs ol ut e val ue
whi l e t he hi ghes t abs ol ut e val ue was obs e r ve d under a ' 20% whi t e' scr een. Ove r most of
t he ni ght s t he var i at i ons in net r adi at i on under t he scr eens wer e s i mi l ar to t hose obs e r ve d
on 1 7 / 1 / 9 3 .
The l ar gest r adi at i ve f l ux t owar ds t he s ky is at about 17:00 h and i t decr eas es wi t h
t i me. The sun set s by 17: 00 h but t he gr ound i s st i l l war m and i t l oses a l ar ge amount of
heat by r adi at i on. Cons e que nt l y i t s t emper at ur e dr ops r esul t i ng a decr eas e in its r adi at i ve
282 M. Teitel et al. /Agricultural and Forest Meteorology 81 (1996) 273-286
-15 [ ]
] X
- 2 0 ¢ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . : . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ~ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ~ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ~ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . = . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. - - - ~ 5 ............... - ............... i . . . . + ............. , ~ .......... ~ ............ ~ .............. ~ ............. ~ .............. ? ........... × ...............
~ + i + i i
- 35 ................................... } ............... t~ .............. ~.
E 4- i i
i ÷ 4- ÷ S ÷
a i ÷ i
1~ - 40 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . i ................................... ~ ................................................................................................... ~ ...................................
. ~ + 1 " . i
- 4 5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . i . . . . . . . . . . . . . , . . . . . . . . . . . . m . . . . . . . . . . . . . . . . . . . . . . . . . . . ' i .............................. " . . . . . . . . . . . . m . . . . . . . . . . . . . . ~ . . . . . . . . . . . . . . " . . . . . . . . . . . . . . .
Z - 50 .................................................................... ~ ................................... i .................................. " .................................................................
i
- 5 5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . , . . . . . . . . . . . . . - i i . . . . . . . . . . . . . . . . ~ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ! . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . r . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . - . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
i i i i !
- - 60 , I , I , i , i , ,
14/01 16/01 18/01 20/ 01 22/ 01 24/ 01 26/ 01
Dat e
Fig. 6. Average, night-time (18:00 h to 06:00 h) net radiation for the period 14-26 January 1993 under 4 types
of screens. (×) 50% aluminized; ([]) 40% black; (+) 30% black; ( • ) 20% white.
heat loss. This process apparently results an exponential decrease of net radiation with
respect to time.
Average net radiation over night time (between 18:00 h and 06:00 h) was calculated
from the experimental data for the different types of screens and is presented in Fig. 6.
The values for the ' 20% white' , ' 30% black' , ' 40% black' and ' 50% al umi ni zed'
screens were - 45, - 35, - 28, and - 25 Wm 2, respectively. Smaller values were
obtained on 14 and 15 January 1993, apparently because the sky was cloudy. This
assumption is supported by the reduction in day time radiation that was observed on the
13/ 1/ 93 (data not shown).
Temperatures of the ground, the upper leaves and the air are shown in Fig. 7(a) and
(b), for the two most efficient frost protective screens ' 50% black' and ' 50% alu-
mi ni zed' , respectively. The data was obtained on the 26/ 1/ 94 and presents typical
temperature variations that were measured during the coldest week of winter 1993/ 1994.
Note that during that winter, the ambi ent temperatures were higher than zero. The lowest
ambient temperature was 4°C and therefore no frost damage could be observed. The
figures suggest that during the night (starting at about 17:00 h) the temperature of the
ground, air and leaves start to drop and reached a mi ni mum level at about 06:00
h- 06: 30 h. During the morni ng (between 06:30 h and 08:30 h) the average change of
temperature of the ground with respect to time, was less steep than those of the air and
the leaves. The moderate change of ground temperature is due to the larger thermal mass
of the ground and the rapid change in heat flux from the sun during the morning. On the
M. Teitel et al. /Agricultural and Forest Meteorology 81 (1996) 273-286 283
18
14
~ tz
~ 10-
6
4 i i i i i ~ i i i
14: oo '18:'oo ' 221oo' 02: 00 ' o81oo
T i me ( h )
' 10: 00' ' ' 14: 00
18
- - . 14
6
4 i i ~ i i i i i D
1 4 : o o l s : ' o o ' z z ; o o ' ' o z ; o o ' ' o 8 ; o o ' l o : b o
T i me ( h )
I 1
14:00
Fig. 7. (a) Temperatures under a '50% black' screen. ( - - ) ground; ( D) ambient; ( • ) air; ( + ) leaf. (b)
Temperatures under a '50% aluminized' screen. ( - - ) ground; (13) ambient; ( • ) air; ( + ) leaf.
284 M. Teitel et a l . / Agri cul t ural and Forest Met eorol ogy 81 (1996) 2 7 3 - 2 8 6
ot her hand, duri ng the night the change in heat flux is much sl ower than in the mor ni ng,
resul t i ng in si mi l ar over ni ght average t emperat ure drop with respect to time in the
ground, in the air and in the leaves.
Whi l e the air and gr ound t emperat ures under the ' 50% bl ack' and ' 50% al umi ni zed'
were appr oxi mat el y equal duri ng winter 1993/ 1994, the t emperat ure of the upper leaves
under these screens appears to be different. The t emperat ure of an upper l eaf under the
' 50% al umi ni zed' is hi gher than that under the ' 50% bl ack' by about 0. 8- 1° C. The
hi gher l eaf t emperat ure is due to better prot ect i on, f r om radi at i ve heat loss, of the
al umi ni zed screen. Under bot h screens the t emperat ure of the leaves was l ower than air
t emperat ure. Thi s behavi or was obser ved under all tested screens. Act ual l y such a
behavi or was expect ed since the leaves l oose heat by l ong wave radiation and t herefore
cool to a l ower t emperat ure than the air.
Since the accurat e measur ement s of l eaf t emperat ure over l ong peri ods of time is
di ffi cul t to accompl i sh wi t h t her mocoupl es, because of the injury caused to the l eaf due
to t her mocoupl e insertion, an at t empt is made here to est i mat e the t emperat ure of an
upper l eaf f r om convent i onal measur ement s of air t emperat ure, wi nd vel oci t y and net
radiation. As s umi ng that the t emperat ure vari at i ons wi t h respect to time are sl ow and
that the heat st orage in the leaf, the transpiration f r om it and the net radiation f r om the
l ower side of the l eaf are negl i gi bl e duri ng the night, the ener gy bal ance equat i on on a
l eaf reduces to:
2h(T~ - TI) = N ( 5)
where h ( Wm - 2 K - 1) is the heat t ransfer coef f i ci ent f r om the leaf, and N ( Wm - 2 ) is
the net radi at i on f r om the leaf. Not e that the net radi at i on f r om the l ower surface of the
upper l eaf is negl i gi bl e since it exchanges heat wi t h ot her l eaves beneat h it whi ch are at
appr oxi mat el y the same t emperat ure as that of the upper leaf.
The heat t ransfer coef f i ci ent h is gi ven (St anghel l i ni , 1987) by:
p C p ( l l T , - Tal + 207u2) °25
h
1174/ °5 ( 6)
where p (kg m- 3) is the densi t y of the air; Cp (J k g - i K- l) is the specific heat of the
air; l ( cm) is a charact eri st i c l engt h of the l eaf and u ( cm s - ~) is the vel oci t y of the air.
Eq. (6) was deri ved for the case of mi xed (forced as well as free) convect i on and it takes
into account fi ndi ngs that a fl uct uat i ng fl ow can expand heat t ransfer from surfaces
(St anghel l i ni , 1987, 1993)
As s umi ng that the net radi at i on f r om the l eaf is equal to t hat measur ed by the net
radi omet ers t hat were posi t i oned about 20 cm above the plants (since the l ong wave
spectral charact eri st i cs of the gr ound and the leaves are appr oxi mat el y equal) and usi ng
the experi ment al dat a of air t emperat ure and wi nd vel oci t y we cal cul at ed using Eqs. (5)
and (6) the t emperat ure of an upper l eaf under a ' 50% al umi ni zed' screen. The
cal cul at ed result t oget her with the experi ment al dat a are shown in Fig. 8. The agr eement
bet ween the experi ment al dat a and the cal cul at i ons is good resulting an average
devi at i on of about 0.2°C. Si nce the devi at i on duri ng ni ght bet ween cal cul at ed and
measur ed l eaf t emperat ures is bot h posi t i ve and negat i ve, absol ut e values are used f or
M. Teitel et al. / Agricultural and Forest Meteorology 81 (1996) 273-286 285
10
9 -
g7
4
19:00 23: 00 03: 00
21:00 01:00 05: 00
T i me ( h)
2
! ~ -1.B
-1.8
/ /
-1.4
• - 1 . 2 o
- i o
. . - 4
. * . J
-0.B t>
0J
-0.6
-0. 4
~ -O.2
, 0
07:00
09: 00
Fig. 8. Temperature of an upper leaf, T t, under a '50% aluminized' screen. ( A) air temperature; (11)
calculated, Eqs. (5) and (6); ( +) experimental data; ( [ ] ) deviation. Calculated and experimental data refer to
leaf temperatures.
cal cul at i ng t he aver age devi at i on. Not e t hat Fi g. 8 pr es ent s t he abs ol ut e val ues of t he
devi at i on. The aver age devi at i ons f or t he ' 20% whi t e' , ' 30% bl a c k' and ' 50% bl a c k'
scr eens wer e 0.2, 0. 4 and 0. 5°C, r es pect i vel y.
Thes e r esul t s s ugges t t hat l eaf t emper at ur e was act ual l y l ower by about 0 . 5 - 1 ° C t han
ai r t emper at ur e. Thi s is in a gr e e me nt wi t h t he r esul t s of Shaw ( 1954) , who s howe d t hat
mi n i mu m ai r t emper at ur es pr ovi de l i t t l e i nf or mat i on on t he t emper at ur e of t he l eaves
dur i ng frost . The t r i gger f or oper at i ng f r ost pr ot ect i ve means shoul d, t her ef or e, be t he
l eaf t emper at ur e and not t he ai r t emper at ur e whi ch i s us ual l y used at pr esent .
6. Co n c l u s i o n s
Shadi ng scr eens s t r et ched hor i zont al l y over t he cr ops can r educe t he r i sk of f r ost
da ma ge ; a l umi ni z e d scr eens wer e f ound to be t he mos t ef f ect i ve. The scr eens shoul d
have hi gh shadi ng per cent age. It s houl d be emphas i zed, however , t hat t he hi gh s hadi ng
per cent age o f t he scr een r educes t he amount of l i ght t hat r eaches t he pl ant s dur i ng t he
day and t hus af f ect s t hei r gr owt h and yi el d. It is r e c omme nde d t hat t hese scr eens shoul d
be st r et ched over t he cr ops onl y when f r ost is expect ed.
A mode l was de vi s e d f or cal cul at i ng t he decr eas e i n net r adi at i on due to t he pr es ence
of a s hadi ng scr een. The mode l s ugges t s t hat net r adi at i on under a scr een is de pe nde nt
286 M. Teitel et al. /Agricultural and Forest Meteorology 81 (1996) 273-286
on t he s ha di ng pe r c e nt a ge o f t he s cr een, on t he t r ans f er f a c t or o f t he s c r e e n ma t e r i a l
/3 = 0. 5 E s + ~-s and on t he r at i o b e t we e n s c r e e n ar ea and t he gr ound ar ea be ne a t h it. Th e
e f f e c t i v e n e s s o f a hor i z ont a l s c r e e n a ppe a r s t o be h i g h e r t han t hat o f a l ow t unnel
f o r me d by s pr e a di ng a s c r e e n o v e r t he pl ant s.
Th e t e mp e r a t u r e o f t he l e a ve s , unde r al l t ypes o f s cr eens , dr ops at ni ght t o l o we r
l e ve l s t han ai r t e mpe r a t ur e and, t he r e f or e , t he c ont r ol o f any f r ost p r o t e c t i v e - d e v i c e s in
t he c a s e o f s e v e r e f r ost s houl d be t r i g g e r e d by l e a f t e mpe r a t ur e and not by ai r
t e mpe r a t ur e .
Acknowledgements
Th e a ut hor s ar e gr at ef ul t o N. Le v a v , A. Le v i and J. Kat z f or t hei r he l p i n c o n d u c t i n g
t he e x p e r i me n t s . Spe c i a l t hanks ar e due t o Dr . I. Se ga l and M. Ba r a k f or he l pf ul
d i s c u s s i o n s and t e c hni c a l s uppor t . We wo u l d al s o l i ke t o t hank B. Ga ml i e l , R. Gol a n, R.
Ch e n and E. Me t z r a f i f or p r o v i d i n g t he a r e a f or e x p e r i me n t s and f or t he i r he l p
t h r o u g h o u t t he e xpe r i me nt . Spe c i a l t ha nks ar e al s o t o Pol ys a c k- pl a s t i c i ndus t r i es , f or
p r o v i d i n g t he s ha di ng s c r e e ns and i n f o r ma t i o n a bout t hei r r a d i o me t r i c pr ope r t i e s .
References
Amsen, M.G., 1975. Heat radiation from a glasshouse crop canopy at night. Acta Hortic., 46: 139-145.
Bailey, B.J., 1981. The reduction of thermal radiation in glasshouses by thermal screens. J. Agric. Eng. Res.,
26: 215-224.
Brooks, F.A., 1961. Frost control. In: C.B. Richey, P. Jacobson and C.W. Hall (Editors), Agricultural
Engineers' Handbook. McGraw-Hill, New York.
Critchfield, H.J., 1966. General Climatology, Second edition. Prentice-Hall, Englewood Cliffs, NJ.
Gat, Z. and Karni, O., 1993. An agroclimate review of winter 1993. R&D Report of the Negev and Arava. pp.
109-117 (in Hebrew).
Love, T.M., 1968. Radiative Heat Transfer. Charles E. Merrill Books, Inc.
Oke, T.R., 1987. Boundary Layer Climates, Second edition, Routledge, London.
Rosenberg, N.J., Blad, B.L. and Verma, S.B., 1983. The Biological Environment, Second edition. Wiley
Interscience, New York.
Shaw, R.H., 1954. Leaf and air temperatures under freezing conditions. Plant Physiol., 29: 102-104.
Stanghellini, C., 1987. Transpiration of greenhouse crops: an aid to climate management. Ph.D. Thesis. IMAG,
Wageningen.
Stanghellini, C., 1993. Mixed convection above greenhouse crop canopies. Agric. For. Meteorol., 66:
111-117.
Teitel, M. and Segal, I., 1995. Thermal net radiation under shading screens. J. Agric. Eng. Res., 61: 19-26.