Journal of Agricultural Chemistry and Environment, 2014, 3, 57-65
Published Online April 2014 in SciRes.
How to cite this paper: El-Shirbeny, M.A., Ali, A.-E.M. and Saleh, N.H. (2014) Crop Water Requirements in Egypt Using Re-
mote Sensing Techniques. Journal of Agricultural Chemistry and Environment, 3, 57-65.
Crop Water Requirements in Egypt Using
Remote Sensing Techniques
Mohammed A. El-Shirbeny1*, Abd-Elraouf M. Ali1, Nasser H. Saleh1
National Authority for Remote Sensing and Space Sciences (Egypt) 23 Joseph Tito St., El-Nozha El-Gedida,
Cairo, Egypt
Email: *
Received Dec emb er 2013
The common Soil in Egypt is clay soil so common irrigation system is tradition surface irrigation
with 60% irrigation efficiency. Agricultural sector consumes more than 80% of water resources
under surface irrigation (tradition methods). In arid and semi-arid regions consumptive use is the
best index for irrigation requirements. A large part of the irrigation water applied to farm land is
consumed by Evapotranspiration (ET). Irrigation water consumption under each of the physical
and climatic conditions for large scale will be easier with remote sensing techniques. In Egypt,
Agricultural cycle is often tow agricultural seasons yearly; summer and winter. Common summer
crops are Maize, Rice and Cotton while common winter crops are Clover and Wheat. Landsat8
bands 4 and 5 provide Red (R) and Near Infra-Red (NIR) measurements and it used to calculate
Normalized Deference Vegetation Index (NDVI) and monitoring cultivated areas. The cultivated
land area was 3,27 7,311 ha in August 2013. In this paper Kc = 2 * NDVI 0.2 represents the rela-
tion between crop coefficient (Kc) and NDVI. Kc and Reference evapotranspiration (ETo) used to
estimate ETc in Egypt. The main objective of this paper is studying the potential crop Evapotrans-
piration in Egypt using remote sensing techniques.
Normalized Deference Vegetation Index (NDVI); Reference Evapotranspiration (ETo);
Landsat8; Crop Coefficient (Kc); Crop Evapotranspiration (ETc) and Arid Region
1. Introduction
Limited water resources and scarcity of water in Egypt is the main challenge for agricultural horizontal expand-
ing policies and strategies. At the same time over population is rapidly increasing and agricultural land decreased.
As a result, quantities and qualities of food decrease. To solve this problem adopted policies of horizontal and
vertical expanding in agricultural lands and activities had been considered based saving irrigation water to cul-
tivate another area. Crop water requirements must be estimated accurately, improving irrigation efficiencies and
applying high efficiency irrigation method as localized irrigation (drip irrigation efficiency 85% - 95%), using
scheduling irrigation, cultivate tolerated crops to drought, and cultivate short period verities. Information about
Corresponding author.
M. A. El-Shirbeny et al.
crop evapotranspiration (ETc) or consumptive water use is significant for water resources p lanning and ir rigation
management. Crop water requirements have to estimate with high level of certainty to save water and to max-
imize water unit uses. Crop evapotranspiration represent crop water requirements in consideration water in-
volved in plant tissue structure representing about 1% or less. Reference evapotranspiration (ETo) is a key
process in land surface studies. It mainly depends on water availability and incoming solar radiation and then re-
flects the interactions between surface water processes and climate [1]. ET c, which is t he c onsu mpti ve water us e
of a field situation where the soil is not under moisture stress, can be estimated using FAO-P e nman-Monteith
Determination of ET c at farm level has traditionally been made on the basis of a so called two steps approach.
The evapotranspiration of a reference standard crop (ETo) is first estimated on the b asis of a site meteorolo gical
variables. A semi-empirical coefficient (crop coefficient, kc) is then applied to take into account all the other
crop and environmental factors [2]. The standardized FAO56 Penman-Monteith model, which has been the most
reasonable method in both humid and arid climatic conditio ns, pr ovides ET o estimates for p lanning and efficie nt
use of agricultural water resources [3] [4] and [5]. To estimate crop evapotranspiratio n follow equation will be
used: -
c oc
where; ETc is crop evapotranspir ation (mm/day), ETo is r eference evapotranspiration (mm/day), Kc is crop coef-
ficient. ET o is defined as the rate of evapotranspiration from a hypothetical reference crop with an assumed crop
height of 0.12 m, a fixed surface resistance of 70 s/ m and albedo of 0.23 closely resembling the evapotranspira-
tion fr o m an e xte nsive surfa ce of gre en gr as s of unifo rm he ight , act ive ly gro wing, well wa tere d, and c omplet el y
shading the ground. A standardization of this method has been proposed by the Food and Agriculture Organiza-
tion [6]. N ume ro us weather variables such as air temperature, relative humidity, wind speed and solar radiation
required to estimate ETo. Consequently, ETo is often estimated by means of empirical equations based on air
temperature, r e la tive humidity, extraterr e str ia l radiation and/or precipitation [7]-[9].
( )( )
( )
0.408 273
ET =1 0.34
n sa
R Gue e
∆ −+−
∆+ +
where; ETo, reference evapotranspiration [mm day1], Rn, net radiation at the crop surface [MJ·m2·day1], G,
soil heat flux density [MJ·m2·day1], T, mean daily air temperature at 2 m height [˚C], u2, wind speed at 2 m
height [m·s1], es, saturation vapour pressure [kPa], ea, actual vapour pressure [kPa], es - ea, saturation vapour
pressure deficit [kPa], slope vapour pr essure curve [kPa·˚C1], psychrometric constant [kPa ·˚C 1].
The crop coefficient coefficient (Kc) is defined as the ratio o f crop p otential evapotr anspiratio n (ET c) to a ref-
erence evapotranspiration (ETo). It is affected by the local climate conditions, crop characteristics, length of
growing season, soil moisture and the time of planting [10] and [6]. ET c can be obtained from ETo using a
stage -dependent crop coefficient.
Vegetation indices (VIs) were first developed in the 1970s to monitor terrestrial landscapes by satellite sen-
sors and have been highly successful in assessing vegetation condition, foliage, cover, phenology, and processes
related to the fraction of photosynthetically active radiation absorbed by a canopy [11] and [12] [13] reported
that satellite-based remote sensing is a robust, economic and efficient tool for estimating actual ET and devel-
oping crop coefficient (Kc) curves. This technique can cover hundreds of sampled fields at a time so that large
populations of ETo and Kc can be used to develop representative mean values. They used empirical equation for
soil adjusted vege tat ion inde x (SAVI) to get cro p coeffici ent through the regressi on a nalys is.
Ka b=∗+
where: SAVI, is soil adjusted vegetation index, a, and b, can be determined by regression analysis.
The NDVI transformation is computed as the ratio of the measured intensities in the red (R) and near infrared
(NIR) spectral bands using the following formula:
() ()
The resulting ind ex va lue i s s ensitive to the presence of vegetation o n the Earths land surface and can be used
to address issues of vegetation type, amount, and condition. Many satellites have sensors that measure the red
M. A. El-Shirbeny et al.
and near-infrared spectral bands, and many variations on the NDVI exist. The sensor that supplies one of the
most widely used NDVI products is on board the Landsat8 with channels in the red (Band 4) and near infrared
(Band 5).
[14] developed a temporal upscaling scheme using satellite -derived instantaneous estimates of ET to pro duce
a daily sum ET averaged over an 8-da y inter val.
[15] found that the sensitivity of NDVI to chlorophyll concentration varied depending on the choice of visible
band used in the calculations. The visible band chosen, therefore, significantly changed the correlation between
the NDVI and canopy properties. They also found that the NDVI tended to saturate as LAI increased. Satellite
maps o f vege tatio n sho w the densit y of pl ant growth. The most co mmo n measurement is called the Normalized
Difference Vegetation Index (NDVI). Very low values of NDVI (0.1 and below) correspond to arid areas of
roc k, sand , o r sno w. Mod er ate va lue s rep res ent s hrub a nd grassl and (0 .2 t o 0.3) , while hi gh val ues i ndi cate te m-
perate and tropical rainforests (0.6 to 0.8). The main objective of this study is studying the potential crop Evapo-
trans piratio n in E gypt usi ng remote s ensing techniques.
2. Materials and Methods
2.1. Study Area
Nile valley and Delta, from Aswan (in the S out h) to Mediterranean Sea (in the North) (Figure 1).
2.2. Remote Sensing Data Availability
Remote sensing provides spatial coverage by measurement of reflected and emitted electromagnetic radiations,
across a wide range of wavebands, from the earth’s surface and surrounding atmosphere. That is mean remote
sensing is the act of collecting data without physically contacting. Landsat8 data collected around 10 a.m. local
time with 30 meter ground resolution, during Aug. 2013 were used in the current stud y. Landsat8 satellite data
were used to calculate NDVI. T he study area covered by 11 images (Table 1).
Figure 1. Study area location.
M. A. El-Shirbeny et al.
Table 1. Used Lan dsat8 data to co vering study area.
NO Path Rew Date NO Path Rew Date
1 174 043 5 Aug 2013 7 176 0 41 03 Aug 2013
2 175 042 12 Aug 2013 8 177 0 38 10 Au g 2013
3 175 043 12 Aug 2013 9 177 0 39 10 Au g 2013
4 176 038 03 Aug 2013 10 177 0 40 10 Aug 2013
5 176 039 03 Aug 2013 11 178 0 39 01 Aug 2013
6 176 040 03 Aug 2013
2.3 NDVI and Kc
The relation between Kc and NDVI is clear. Landsat8 bands 4 and 5 provide R and NIR measurements and
therefore can be used to generate NDVI data with the following formula:
() ()
NDVIBand 5Band 4Band 5Band 4=−+
Kc used wit h ET o to esti mate ETc. Kc is a dimensionless number (usually between 0.1 and 1.2) that is multip-
lied by the ETo value to calculate (ETc). The resulting ETc can be used to help an irrigation manager schedule
when irrigation should occur and how much water should be put back into the soil. In this paper, the relation
between Kc and NDVI represented by Equation (6).
( )
1.2 NDVI 0.2
K= −
2.4. Classification
Landsat8 data were used to classify NDVI using unsupervised classification. Initial unsupervised classification
(K-means) was applied which is an automated cluster analysis technique that uses a minimum spectral distance
cluster algor ithm to assign a p ixel to a cluster o f pixels with similar a ttribute [16].
2.5. Penm a n-Montieth Method
ETo was calculated from the meteorological data using the Penman-Montieth formula. This formula applied by
empirical method to calculate ETo, which adjusted by crop coefficient to estimate ETc. The meteorological pa-
rameters used in this equation were taken from Alex., Tanta, Wadi El-Natron, Cairo, I smailia, Po rtsaid , El-Seuz,
Helwan, M in ya, Asyut, Sout h of va lley, Luxor and Aswan st atio ns (Figure 2) sho ws the distributio n of stations.
3. Results and Discussion
3.1. Normalized Deference Vegetation Index (NDVI)
NDVI calculated from Red and NIR bands in the satellite data. The NDVI equation produces values in the range
of 1.0 to 1.0, where vegetated areas typically have values greater than 0.2 and less values indicate non-vegetated
surface features such as water, barren, ice, snow, or clouds. NDVI var y according to crop age, planting density
and chlorop hyll activity. It seems like Kc var ying fro m planti ng to senesce nce. NDVI used as input i n Equatio n
(6) to e stimate Kc from satellite data. Fig ure 3 and Figure 4 show NDVI and Kc of Nile valley and delta.
3.2 Reference Evapotranspiration (ETo)
To solve the problem of water limitation and scarcity in Egypt, adopted policies of horizontal and vertical ex-
panding in agric ultural land s and activities had been con sidere d based savin g irrigation water to culti vate anoth-
er area. Agriculture is responsible for 70% of all water use globally and water use efficiency in this sector is
ver y low, not exceed ing 45 % [17]. ET o affected directly with climate and it depend completely on meteorologi-
cal data (Temperature, Relative humidity, Wind speed and Solar radiation). When vegetation covers soil 100%,
all water losses are transpiration. ETo was estimated from meteorological data according to Penman-Montieth
method. Mean ETo was 8.3 (mm). Maximum ETo was 10.61 (mm) at Aswan. Minimum ETo was 6.1 (mm) at
M. A. El-Shirbeny et al.
Figure 2. The distribution of meteorological stations.
Figure 3. NDVI distribution.
M. A. El-Shirbeny et al.
Figure 4. Kc distribution..
Figure 4. Kc distribution..
M. A. El-Shirbeny et al.
Alexandria. Comparison between monthly ETo (Alex.) and monthly ETo (Aswan) was shown in Table 2. In-
crease of ETo was observed in south of Egypt because of increasing of radiation and air temperature and de-
creasing of relative humidity.
3.3. Crop Evapotranspiration (ETc)
Kc used with ETo to estimate ETc according to Equation (1). From Figure 5, it has been observed to increase
ETo in the southern part of Egypt because radiation and air temperature were high and relative humidity was
Table 2. ETo for Alex. at north and Aswan at south (mm/day).
Month ETo (Alex.) ETo (Aswan)
Jan. 20 1 3 2.3 4.5
Feb. 2013 3.6 5.9
Mar. 2013 3.9 7 .7
Apr. 2013 4.5 9.5
May 201 3
5.1 1 0.5
Jun. 2013 6.2 11.7
Jul. 2013 6.3 1 1. 1
Aug. 2013 6.1 10.61
Sep. 2013
4.9 1 0.1
Oct. 2013 3.9 8 .4
Nov. 2013 2.8 6.1
Dec. 2013 2.1 4.7
Averag e
4.3 8 .4
Figure 5. ETc distribution in nile valley and
delta (egypt).
M. A. El-Shirbeny et al.
4. Conclusion
The cultivated areas in Nile valley and delta were about 3.3 Mha. Landsat8 bands 4 and 5 used to calculate
Normalized Deference Vegetation Index (NDVI). Kc = 2 * NDVI 0.2 represent the relation between crop
coefficient (Kc) and NDVI. ETo was calculated from the meteorological data using the Penman-Montieth for-
mula. Kc and ETo used to estimate ETc in Egypt. It has been observed to increase ETo in the southern part of
Egypt because of radiation and air temperature were high and relative humidity was low.
I would like to thank NASA for data availability. I would like to thank my collage at NARSS, Egypt for their
support and encourage me nt .
[1] Sobrino, J.A., Gomez, M., Jimenez-Munoz, J.C. and Olioso, A. (2007) Application of a Simple Algorithm to Estimate
Daily Evapotranspiration from NOAA–AVHRR Images for the Iberian Peninsula. Remote Sensing of Environment,
110, 139-148.
[2] Magliulo, V., d’Andria, R. and Rana, G. (20 03) Use of t he Modified Atmomet er to E stimate Reference Evap otranspi-
ration in Mediterranean En vir onments. Agricultural Water Management, 63, 1-14.
[3] Yin, Y., Wu, S., Zheng, D. an d Y ang, O. (2008) Radiation Calibration of FAO56-Penman-Monteith Model to Estimate
Reference Crop Evapotranspiration in China. Agr i c u l tur a l Water M ana ge m e nt , 95, 77-84.
[4] El-Shirbeny, M.A., Aboelghar, M.A., Arafat, S.M. and El-Gindy, A.-G.M . (20 14) Assessment of the Mutual Impact
between Climate and Vegetation Cover Us ing NOAA-AVHRR and Landsat Data i n Egypt. Arabian Journal of Geos-
ciences, 7, 12 87 -1296 .
[5] Zhao, S.H., Yang, Y.H., Zhang, F., Sui, X.X., Yao, Y.J., Zhao, N. and Li, C.Q. (20 14 ) Rapid Evaluation of Reference
Evapotranspiration in Northern China. Arabian Journal of Geosciences, 1-11.
[6] Allen, R.G., Perrier, L.S., Raes, D. and Smith, M. (1998) Crop Evapotranspiration: Guidelines for Computing Crop
Requirements. Irrigation and Drain age Paper No. 56, FAO, Rome, Italy.
[7] Droogers, P. and Allen, R.G. (2002) Estimating Re ference E vapotr anspiration under Inaccu rate Data Condit ions. Irri-
gation and Drainage Systems, 16, 33-45.
[8] Hargreaves, G.H. and Samani, Z.A. (1985) Reference Crop Evapotranspiration from Temperature. Applied Engineer-
ing in Agri c ulture , 1, 96-99.
[9] Popova, Z., Kercheva, M. and Pereira, L.S. (2005) Validation of the FAO Methodology for Computing ETo with Li-
mited Data. ICID 21st European Regional Conference, Fran kfur t, Slubice.
[10] Doorenbos, J. and Pruitt, W.O. (1977) Crop Water Req uirement: Food and Agriculture Organization of the United Na-
tions. FAO Irrigation and Drainage Paper 24, Rome, 144.
[11] Huete, A., Didan, K., Van Leeuw en, W., Miura, T. and Glenn, E. (2008) MODIS V egetation Indi ces. In Land Remote
Sensing and Global Environmental Change, NASA’s Earth Observing System and the Science of ASTER and MODIS .
[12] Edward, P.G., Huete, A.R., Nagler, P.L. and Nelson, S.G. (2008) Relationship between Remotely Sensed Vegetation
Indices, Canopy Attributes and Plant Physiological Processes: What Vegetation Indices Can and Cannot Tell Us about
the Landscape. Sensors, 8, 2136-2160. Environment, 48, 119-126.
[13] Ta sumi, M. and Allen, R.G. (2007) Satellite-Based ET Mapping to Assess Variation in ET with Timing of Crop De-
velopment. Agricultural Water Management, 88, 54-62.
[14] Ryua, S.Y., Baldocchi, D.D., Black, T.A., Dettoc, M., Lawd, B.E., Leuninge, R., Miyataf, A., Reichsteing, M., Var-
gash, R. , Ammanni, C ., Berin ger, J., Flanagan k, L.B., Gul, L., Hutleym, L.B., Kimn, J., McCaugheyo, H., Moorsp, E.
J., Rambal, S. and Vesalar, T. (2012) On the Temporal Upscaling of Evapotranspiration from Instantaneous Remote
Sensing Measur ement s to 8-Day Mean Dai ly Sums. Agricultural and Forest Meteorology, 152, 212 -222.
[15] Yoder, B.J. and Waring, R.H. (1994) The Normalized Difference Vegetation Index of Small Douglas-Fir Canopies
with Varying Chlorophyll Concentrations. Re m ot e Se ns ing of Env ironm e nt , 49, 81-91.
[16] Campbell, J.B. (1996) Introduction to Remote Sensing. Taylor Francis, London.
M. A. El-Shirbeny et al.
[17] Zhang, X., Friedl, M.A., Schaaf, C.B. and Strahler, A.H. (2004) Climate Controls on Vegetati on P heno lo gical P at tern s
in Northe r n Mi d and High Lati tudes Inferred from MODIS Data, Glob. Change, 10, 1133-1145.