Estimation of Crop Evapotranspiration Using Satellite ...downloads.hindawi.com/journals/amete/2018/4525021.pdf · Estimation of Crop Evapotranspiration Using Satellite Remote Sensing-Based

Research ArticleEstimation of Crop Evapotranspiration Using Satellite RemoteSensing-Based Vegetation Index

Arturo Reyes-Gonzaacutelez 12 Jeppe Kjaersgaard3 Todd Trooien1

Christopher Hay4 and Laurent Ahiablame1

1Department of Agricultural and Biosystems Engineering South Dakota State University Brookings SD USA2Instituto Nacional de Investigaciones Agrıcolas Forestales y Pecuarias (INIFAP) Blvd Prof Jose Santos ValdezNo 1200 Pte Col Centro Matamoros COAH Mexico3Minnesota Department of Agriculture St Paul MN USA4Iowa Soybean Association Ankeny IA USA

Correspondence should be addressed to Arturo Reyes-Gonzalez reyesarturoinifapgobmx

Received 25 August 2017 Accepted 26 December 2017 Published 1 February 2018

Academic Editor Jan Friesen

Copyright copy 2018 Arturo Reyes-Gonzalez et al This is an open access article distributed under the Creative Commons AttributionLicense which permits unrestricted use distribution and reproduction in any medium provided the original work is properlycited

Irrigation water is limited and scarce in many areas of the world including Comarca Lagunera Mexico Thus better estimationsof irrigation water requirements are essential to conserve water The general objective was to estimate crop water demands orcrop evapotranspiration (ETc) at different scales using satellite remote sensing-based vegetation index The study was carried outin northern Mexico (Comarca Lagunera) during four growing seasons Six eleven three and seven clear Landsat images wereacquired for 2013 2014 2015 and 2016 respectively for the analysis The results showed that ETc was low at initial and earlydevelopment stages while ETc was high during mid-season and harvest stages These results are not new but give us confidence inthe rest of our ETc results Daily ETc maps helped to explain the variability of crop water use during the growing season Based onthe results we can conclude that ETc maps developed from remotely sensed multispectral vegetation indices are a useful tool forquantifying crop water consumption at regional and field scales Using ETc maps at the field scale farmers can supply appropriateamounts of irrigation water corresponding to each growth stage leading to water conservation

1 Introduction

Irrigation water is limited and scarce in many areas of theworld Agriculture is the major consumer of fresh water[1 2] but it is not necessarily used efficiently due to farmerssupplying more water than is consumed by the crop Thusbetter estimation of irrigation water requirements is essentialto use water efficiently so water is available for use in thefuture To achieve water conservation it is necessary thatfarmers adopt new technologies for estimating crop waterdemands more accurately

Crop evapotranspiration (ETc) represents crop waterrequirement and is affected by weather and actual cropconditions [3 4] A useful method to estimate ETc or cropwater requirements is to multiply reference evapotranspira-tion (ETr) by a crop coefficient (119870c) (see (1)) ETr is estimatedbased onmeteorological information (eg solar radiation air

temperature wind and air vapor pressure deficit) froma localweather station The Penman-Monteith equation has beenadvanced as the standard method for estimating referenceET [5 6] 119870c is typically taken from literature values and isaffected by crop variety and growth stage [5ndash7] ETc has beenestimated or measured using other methods for exampleweighing lysimeters evaporation pan soil water balanceatmometer Bowen Ratio Energy Balance System (BREBS)and Eddy Covariance (EC) However these methods arerecognized as the point-based measurements Satellite-basedremote sensing is an alternative to estimate crop waterrequirement and its spatial and temporal distribution on afield-by-field basis at a regional scale These remote sensing-based methods have been shown to be accurate [8ndash11]

ETc = ETr times 119870c (1)

HindawiAdvances in MeteorologyVolume 2018 Article ID 4525021 12 pageshttpsdoiorg10115520184525021

2 Advances in Meteorology

Torreoacuten Coah

0 5 10 20(Km)

Meacutexico

1000 km500 mi

26∘0 0

N

25∘45 0

N

25∘30 0

N

26∘0 0

N

25∘45 0

N

25∘30 0

N

103∘400W 103∘200W 103∘00W

103∘400W 103∘200W 103∘00W

(b)

(a)

Figure 1 Location of the study area at northern Mexico (a) The subset of the area of interest Landsat with false color composite (bands 43 2) the yellow triangles denote five locations where we selected the corn fields and the white star indicates weather station (b)

Remote sensing is a technology that can estimate ETcat regional and local scale in less time and with less cost[9 10] Remote sensing can also estimate crop coefficientsbased on spectral reflectance of vegetation indices (VIs) [512] The normalized difference vegetation index (NDVI) isthe most common VIs [13] NDVI takes into account thereflectance of red and near-infrared wavebands [14] wherered waveband is strongly absorbed by chlorophyll in leaves ofthe top layers while near-infrared wavebands is reflected bythemesophyll structure in leaves penetrating into deeper leaflayers in a healthy vegetation [15 16]High values ofNDVI arerelated to healthy and dense vegetation which presents highreflectance values in the NIR waveband and low reflectancevalues in the red waveband [17] Crop coefficients generatedfrom VIs determine ETc better than a tabulated119870c because itrepresents the actual crop growth conditions and capture thespatial variability among different fields [2 18 19]

Several studies have usedmultispectral vegetation indicesderived from remote sensing to estimate119870c values on agricul-tural crops including corn crop for example in [20ndash31] Cropcoefficients derived from remotely sensed vegetation indexalso have used to generate local and regional ETc maps [2 32ndash34] however in northern Mexico ETc maps using satelliteremote sensing-based vegetation index remain unexplored

The objectives of this study were to (1) calculate NDVIvalues for each corn field for each growing season (2)develop a simple linear regression model between NDVI

derived from satellite-based remote sensing and tabulated119870cobtained of alfalfa-based crop coefficient fromASCEManual70 (3) generate119870c maps using the linear regression equationobtained between NDVI and 119870c values and (4) create ETcmaps with high spatial resolution at regional and field scales

2 Materials and Methods

21 Study Area The study was carried out in northernMexico (Comarca Lagunera) during four growing seasonsComarca Lagunera had an average latitude of 25∘401015840N andlongitude of 103∘181015840Wand elevation of 1115m abovemean sealevel (Figure 1) In Comarca Lagunera forage crops (alfalfacorn sorghum and oat (planted in the winter season))occupiedmore than 75 of the total irrigated area [35] Silagecorn is the most important crop after alfalfa in this regionFive silage corn fields in each growing season were selectedfor NDVI calculations The corn fields were irrigated usingsurface irrigation systems The plant population density was78000 plants haminus1 Silage corn is typically planted from lateMarch to early April and chopped for silage from late July toearly August depending on the crop variety The corn fieldsselected ranged between 10 and 20 hectares in size The soiltexture for this region is clay loam soil The mean annualmaximum temperature is 28∘C minimum 13∘C and mean21∘C [36] The mean annual precipitation is 200mm whilethe annual potential evapotranspiration is 2000mm [37]

Advances in Meteorology 3

Table 1 The year acquisition dates day after planting (DAP) Landsat satellite and pathrow for 2013 2014 2015 and 2016 growing seasons

Year Acquisition Dates DAP Satellite PathRow

2013

April 14 10 Landsat 8 3042



May 16 42 Landsat 8 3042

June 9 66 Landsat 7 3042


2014









July 6 88 Landsat 8 3042



2015April 28 22 Landsat 7 3042



2016








22 Landsat Images Clear sky images from Landsat 7Enhanced Thematic Mapper Plus (ETM+) and Landsat 8Operational Land Imager (OLI) andThermal Infrared Sensor(TIRS) (Path 30 Row 42) were used to estimate NDVI119870c and ETc values The images were downloaded from theUnited States Geological Survey (USGS) EROS DatacenterEROS performed the atmospheric corrections on the imagesAlso the wedge-shaped gaps appearing within the Landsat 7images as a result of the SLC-off issue were removed usingthe Imagine built-in focal analysis tool [11] Six eleven threeand seven clear Landsat images were acquired for 2013 20142015 and 2016 respectively (Table 1)The satellite imageswereprocessed using the Model Maker tool of ERDAS ImagineSoftware

23 Pixel Selection Ten pixels for each corn field and eachseason were selected and extracted from NDVI maps Thepixels were located in the center of each corn field for eachoverpass date during the four growing seasons The samepixels were observed throughout the corn growing seasonWe assumed that the pixels are representative of the entirecorn field All corn fields hadflat terrainThenumber of pixelsper year is presented in Table 2

Table 2 The year and number of pixels selected throughout thegrowing season

Year Number of pixels2013 3002014 5502015 1502016 350

24 NDVI Calculations TheNDVI is the difference betweennear-infrared (NIR) and red waveband reflectances dividedby their sum [13] NDVI values range between minus1 and +1where water presents negative values and dense canopypresents high positive values [17 38 39] The NDVI wascalculated for each overpass date and for each growing seasonusing Model Maker tool of ERDAS Imagine Software asshown in the next equations

For Landsat 7 NDVI was calculated as follows

NDVI =(NIRband 4 minus Redband 3)

(NIRband 4 + Redband 3) (2)


Weather data

NIR and redwavebands

NDVI

Inputs

Calculations

Output

Landsat image

Crop coefficient(Kc)

Crop ET(４＝)

Reference ET(４Ｌ)

Figure 2 Flowchart of crop evapotranspiration estimation using vegetation index method




where NIRband and Redband are the near-infrared and redwavebands respectively

Reyes-Gonzalez et al [40] made some comparisons ofNDVI derived from satellite remote sensing and derived fromGreenSeeker in the Comarca Lagunera Mexico The resultsshowed that the NDVI values derived from satellite werewithin 95 of the in situ values (data not shown)

25 Crop Coefficient (119870119888) Values from Manual 70 The 119870c

values were taken from ASCE Manual 70 (Appendix E) [7]and were adjusted according to different corn growth stagesthroughout the growing season For119870c estimations the ASCEManual 70 divides the growing season into two periodsnamely percent of time from planting to effective cover anddays after effective cover to harvest The effective cover andharvest of corn in our study occurred around 55 and 105DAPrespectively based on the crop phenology

26 Relationship betweenNDVI and119870119888and119870

119888MapsDevelop-

ment The relationships between NDVI derived from Land-sat images and tabulated 119870crsquos values obtained from ASCEManual 70 (Appendix E) [7] corresponding to each satelliteoverpass date for 2013 2014 2015 and 2016 corn growingseasons were established These relationships were used togenerate a single linear regression equation for entire periodof study

27 Reference Evapotranspiration (119864119879119903) Calculations The

meteorological information was taken from an automatedweather station The weather station was located at theNational Institute of Forestry Agriculture and LivestockResearch (INIFAP)Matamoros CoahuilaMexico (Figure 1)

The ETr values were taken from the weather station whereETr was calculated using the Penman-Monteith equation[5 41]

28 Crop Evapotranspiration (119864119879119888)Maps The119870c values from

the 119870c maps were multiplied by ETr (see (1)) to create ETcmaps with high spectral resolution (30m) for 2014 growingseason using Model Maker tool of ERDAS Imagine Softwareand ArcGIS version 1031 The ETc maps are generated tomonitor the spatial distribution and temporal evolution of thecrop water requirements during the growing season

29 Flowchart of ETc Estimation A summary of ETc esti-mation using satellite remote sensing-based vegetation indexis shown in Figure 2 The Landsat images and weather dataare the two major inputs parameters in the vegetation indexmethod

3 Results and Discussion

31 NDVI Curves The NDVI average values (10 pixels) se-lected and extracted from NDVI maps for five corn fieldsand for different corn growing seasons are shown in Figure 3The figures show similar NDVI curves for 2014 and 2016while for 2013 and 2015 the curves are incomplete due to lackof clear sky images during the growing seasons In generalNDVI values at initial stage were low around 015 in earlyApril (DAP sim 8) Then the NDVI values increase as the cropdevelops reaching its maximum value (08) at mid-seasonstage followed by plateau from late May to middle July (DAP55ndash95) At the end of the season the NDVI values are slightlydecreasing (sim07) by the end of July (DAP 105) Severalresearchers reported similar seasonal NDVI curves for corn(eg [17 20 24 30 42ndash47]) All NDVI curves developedby these researchers showed low values at early stages thenincreasing at mid-season stages and then slightly decliningat late stages However Thomason et al [44] reported that


L7L8

15 25 35 45 55 65 75 85 95 1055DAP (2013)

0010203040506070809

ND

VI

Field 1Field 2Field 3

Field 4Field 5

L7 L7 L8L8

0010203040506070809

ND

VI

15 25 35 45 55 65 75 85 95 1055DAP (2014)


Field 4Field 5

L7L8

15 25 35 45 55 65 75 85 95 1055DAP (2016)

0010203040506070809

ND

VI


Field 4Field 5

15 25 35 45 55 65 75 85 95 1055DAP (2015)

0010203040506070809

ND

VI


Field 4Field 5

Figure 3 Seasonal evolution of NDVI at five corn fields for 2013 2014 2015 and 2016 growing seasons in northern Mexico

NDVI curves of forage corn gradually increased from initialto mid-season and then remained constant until the end ofthe season

In this study the NDVI values derived from Landsat 8(L8) were greater than NDVI values derived from Landsat7 (L7) in mid-season (Figure 3 (2014 and 2016)) and earlystages (Figure 3 (2013)) The differences between L8 and L7ranged from 003 to 009 (data not shown)Those differencesare in agreement with values reported by Flood [48] (004)andKe et al [49] (006) but greater than reported byRoy et al[50] (002)The difference between L8 and L7 was because L8has a narrower near-infrared waveband (L7 = 077ndash090 120583mL8 = 085ndash088 120583m) higher signal to noise ratio and higher12-bit radiometric resolution [48 49 51 52] These featuresprovide more precise measurements that are less influencedby atmospheric conditions and more sensitive to surfacereflectance [48 49 52] Although the comparison of NDVIbetween L8 and L7 was not an objective of this study it isimportant tomention that inconsistent or unreliable values ofNDVI can produce poor estimates of crop evapotranspiration[49]

32 Relationship between NDVI and 119870119888 The NDVI values

were taken from NDVI maps generated as an output usingLandsat 7 and Landsat 8 while 119870c values were taken fromASCEManual 70 (Appendix E) table for 2013 2014 2015 and2016 corn growing seasons Figure 4 shows the relationshipbetween NDVI of five corn fields and tabulated 119870c valuesfor four growing seasons Strong relationships were observedfor 2013 and 2015 growing seasons with 1199032 equal to 099whereas for 2014 and 2016 1199032 was equal to 096 The slightly

low values of 1199032 found in 2014 and 2016 seasons probably wasdue to major numbers of NDVI values where some of themwere lower than119870c values especially in development growthstage Similar values of coefficients of determination (099)between NDVI and 119870c for corn crop were found by Rochaet al [53] and Reyes-Gonzalez et al [54] but low coefficientswere reported by Sing and Irmak [45] Kamble et al [30] andToureiro et al [17] who reported values of 1199032 to be equal to083 081 and 082 respectively

The NDVI computed from Landsat images and 119870cobtained from Appendix E of ASCE Manual 70 were used todevelop the linear regression equations for 2013 2014 2015and 2016 growing seasons (Figure 4)Those linear regressionswere similar to the linear equations of corn reported by otherresearchers for example in [20 53ndash55] all these authorsused alfalfa-reference crop coefficient and NDVI values forgenerating linear regression equations

The linear regression equations for the four years werecompared using the 119905-test method to test statistical differencebetween two independent regressions [56] Table 3 shows theresults of all comparisons where all 119905 values were less thantabulated 119905 values which means that there were no statisticaldifferences between linear regression equations Based onthese results all data from the four years were pooled and asingle linear equation was generated (Figure 5) This linearequation was used to create119870c maps for 2014 growing season(Figure 6)

33 119870119888Maps and 119870

119888Curve Previous empirical linear equa-

tion between NDVI and 119870c was used to generate 119870c mapsusing Landsat images processed in ERDAS Imagine Software


2014

02 04 06 08 10NDVI

0

02

04

06

08

1

Kc


Field 4Field 5

2013

02 04 06 08 10NDVI

0

02

04

06

08

1

Kc


Field 4Field 5

2016

02 04 06 08 10NDVI

0

02

04

06

08

1

Kc


Field 4Field 5

2015

02 04 06 08 10NDVI

0

02

04

06

08

1

Kc


Field 4Field 5

Kc = 13301＄６） + 00021R2 = 09875

Kc = 14556＄６） + 00618

R2 = 0998

Kc = 12234＄６） + 00242

R2 = 09586

Kc = 10968＄６） + 01054

R2 = 09563

11 line

11 line 11 line

11 line

Figure 4 Linear relationship between NDVI derived from NDVI maps and119870c from ASCEManual 70 for four growing seasons The dashedline indicates the 1 1 line

Table 3 Comparisons between linear regression equations usingthe 119905-test method

Compared years 119905 value 119905 from table2013 to 2014 114 2162013 to 2015 096 2572013 to 2016 213 2262014 to 2015 153 2222014 to 2016 108 2142015 to 2016 231 244

(Model Maker) for the 2014 growing season Figure 6 showsspatial and temporal variability of 119870c values throughout the2014 growing season The 119870c maps showed low 119870c valuesearly in the growing season (DAP 8) (light blue-green color)and gradually increase at mid-season stage (DAP 56) whereplateau remains until harvest (DAP 105) (brown color)Similar 119870c maps for corn were developed by Sing and Irmak[45] Irmak et al [57] Rocha et al [53] and Reyes-Gonzalezet al [54] who reported maps of daily spatial distribution of119870c for six four seven and four overpass dates respectively

ALL

R2 = 097

Kc = 12358＄６） + 00245

02 04 06 08 10NDVI

0

02

04

06

08

1

Kc


Field 4Field 511 line

Figure 5 Linear relationship betweenNDVI and119870c for all dataThedashed line indicates the 1 1 line

Theweather during our 2014 season resulted in usable imagesfor nearly every satellite overpass on 8-day intervals These


DAP 40DAP 32DAP 8 DAP 24

DAP 48

DAP 88 DAP 96 DAP 104


Kc

00 05 10

Figure 6 Spatial and temporal evolution of 119870c generated with ERDAS Imagine Software (Model Maker) and ArcGIS version 1031 duringthe 2014 growing season in northern Mexico

119870c maps show how the119870c values increase (from 02 to 10) asthe silage crop develops

The 119870c values obtained from 119870c maps based on theaverage of ten selected pixels within each of five corn fieldsfor each overpass date throughout the 2014 growing season isshown in Figure 7 In general the minimum 119870c value (024)was presented in early season while the maximum 119870c value(100) was presented in the mid-season stage The standarddeviation (vertical bars) values of 119870c were lower than 007throughout the growing season (Figure 7) this means thatplanting dates management practice and maturity datesamong corn fields did not affect too much the 119870c valuesduring the season

The relationship between119870c calculated from119870cmaps and119870c from tables is shown in Figure 8 A strong relationshipwas found with 1199032 = 096 This means that 119870c valuesderived from vegetation index (119870c calculated) can be a robustparameter to calculate actual crop evapotranspiration Themain difference between 119870c calculated and 119870c tabulated isthat the 119870c tabulated comes from well-water reference crop(eg alfalfa) whereas 119870c calculated comes from the actualcrop growth conditions where some 119870c values derived from

24 32 40 48 56 64 80 88 96 1048DAP (2014)

00

02

04

06

08

10

12

Kc

Figure 7 Crop coefficient curve of corn throughout the 2014growing season Vertical bars represent standard deviation values

reflectance of vegetation were reduced by soil water contentSimilar results were reported by Singh and Irmak [45] andKamble et al [30] They found that limited moisture contentdecreased119870c valuesThose low119870c values occurred around 15days previous to start the mid-season stage In our study thelittle difference between 119870c calculated and 119870c tabulated was


y = 09684x + 00302

R2 = 09588

02 04 06 08 10 1200Kc table

00

02

04

06

08

10

12

Kc

calc

ulat

ed

11 line

Figure 8 Relationship between 119870c calculated and 119870c tabulated for corn during 2014 growing season The dashed line indicates the 1 1 line


DAP 48



0 4 8４＝ (mmday)

Figure 9 Spatial and temporal ETc maps generatedwith ERDAS Imagine Software (ModelMaker) andArcGIS version 1031 for 2014 growingseason in northern Mexico

found in the development growth stage (DAP 40ndash48) whenlimited soil moisture content was presented

34 ETc Maps and ETc Values ETc maps of 30m resolutionwere generated as an output of 119870c maps multiplied by ETrvalues for corresponding day using ERDAS Imagine Software(Model Maker) for 2014 growing season (Figure 9) The ETcmaps showed low ETc values (20mmdayminus1) (light green

color) at initial stage and high ETc values (80mmdayminus1) (redcolor) at mid-season stage These two seasons were charac-terized because in the initial stage crop needs smaller waterrequirements whereas in the mid-season crop needs higherwater requirements as we can see in the next section TheETc maps created in this study are in agreement with otherresearchers for example in [2 4 58 59] they generated ETcmaps using119870c derived from remote sensing-based vegetation



DAP 48



Corn field

8

4

0

４＝ (mmday)

Figure 10 ETc maps at a field scale (eg silage corn) generated with ERDAS Imagine Software (Model Maker) and ArcGIS version 1031using Landsat 7 and Landsat 8 satellite images for the 2014 growing season The red light green and dark blue colors within the corn field(black rectangle) indicate low medium and high ETc values

indices Other researchers reported that 119870c derived fromcanopy reflectance based vegetation index had the potentialto estimate crop evapotranspiration at regional and fieldscales for example in [17 18 29 55 60 61]

35 ETc Maps at a Field Scale ETc maps at filed scale helpto explain the variability of crop water requirements duringthe growing season as shown in Figure 10 These imagesat a field scale level show the corresponding ETc valuesaccording to each growth stage this indicates that each stagerequires different amount of water throughout the seasonFor example minimum water requirements (20mmdayminus1)are needed at the initial stage whereas maximum waterrequirements are needed atmid-season stage (80mmdayminus1)Understating different crop growth stages and applying theaccurate amount of volumetric water farmers can improvetheir irrigation scheduling improve water management andenhance irrigation water sustainability

Similar ETc maps at a field scale for agricultural cropsincluding corn were reported by Farg et al [32] Zipper andLoheide [62] and Senay et al [63] they reported minimum

and maximum ETc values at different crop growth stageswhere the higher evapotranspiration rates were found at themid-season growth stage and lowest evapotranspiration rateswere found at early growth stage

36 Comparison between ETr and ETc The ETr values weretaken directly from a local weather station while ETc valueswere derived from ETc maps Figure 11 shows the comparisonbetweenETr andETc for 214 growing seasonThis figure illus-trated that the daily ETr values were higher than the daily ETcoutputs at the beginning of the growing season but similarvalues were recorded atmid-season stage In early stage (DAP1ndash20) the ETr values were around 60mmdayminus1 while theETc values were around 20mmdayminus1 In development stage(DAP 20ndash55) the ETr values continue around 6mmdayminus1while ETc values increase from 2 to 6mmdayminus1 In the mid-season stage (DAP 55ndash95) both ETr and ETc values werevery similar around 70mmdayminus1 At the end of the growingseason (DAP95ndash105) the ETr valueswere slightly greater thanETc values by 05mmdayminus1 From early to mid-developmentstage the ETc values were lower than ETr values this means


Water save

0

2

4

6

8ET

(mm

daＳminus1)

15 25 35 45 55 65 75 85 95 1055DAP (2014)

４Ｌ

４＝

Figure 11 Comparison between ETr and ETc for 2014 growingseason in northernMexicoThe grey wide column indicates the timeinterval where producers can save irrigation water

that in those particular stages farmers can save irrigationwater (grey wide column in the graph) because in thosestages the crop needs small water requirements due to thefact that the crop canopy is no yet fully developed In shortthe ETc values from ETc maps could be used by farmers intheir irrigation scheduling programs because it shows whenand how much water is required by the crop during differentgrowth stages

Reyes-Gonzalez et al [54] reported that the farmersshould use ETc instead of ETr for irrigation scheduling in aridand semiarid region where irrigation water is scarce USDA-NASS [64] reported that the farmers in the United Statesuse four primary methods to determine when to irrigatethe first was visual observation of crop condition method(41) the second was the soil feel method (21) the thirdwas personal calendar schedule method (11) and the fourthwas daily crop evapotranspiration method (4) Methodsfor deciding when to irrigate need to be more accuratebecause of the competition for irrigation water increases andits value increases Farmers must use scientific irrigationscheduling methods (eg ET method) instead of empiricalmethods (eg crop condition feel of soil and personalcalendar) to save water and protect the environment The ETmethod is based on climatic demands and is more accuratethan empirical methods for irrigation scheduling [65] ETmaps and atmometers are methods to estimate crop waterrequirements with high accuracy [11]

4 Conclusions

Thegeneral objective of this studywas to estimate crop evapo-transpiration using satellite remote sensing-based vegetationindex in northern Mexico

The relationships between NDVI derived from Landsatimages and tabulated 119870c obtained from ASCE Manual 70(Appendix E) were established for four growing seasonsThese empirical linear regression equations were used togenerate a single linear regression equation

ETc maps were created as an output of 119870c maps mul-tiplied by ETr values The ETc values ranged from 140 to741mmdayminus1 during the period of studyThe results showed

that ETc values were low at initial and early developmentstages while ETc values were high during mid-season andharvest stages Daily ETc maps helped to explain the variabil-ity of crop water use throughout the growing season

Farmers in northern Mexico region use empirical meth-ods in their irrigation scheduling methods The resultsindicate that farmers could reduce their seasonal waterapplication amounts by 18 just by using ETc appropriatelyin their irrigation scheduling methods

The information generated in this study is essential forirrigation scheduling because it shows when and how muchwater is required by the crop during different crop growthstages

According to our results we can conclude that ETc mapsdeveloped from remotely sensed multispectral vegetationindices are a useful tool for quantifying accurate crop waterconsumption or crop evapotranspiration at regional and fieldscales

Conflicts of Interest

The authors declare that they have no conflicts of interest

Acknowledgments

The first author would like to acknowledge the NationalCouncil for Science and Technology of Mexico (CONACYT)and the National Institute of Forestry Agriculture andLivestock Research (INIFAP) for funding his doctoral schol-arship Additional funding is provided by the South DakotaAgricultural Experiment Station and the SouthDakotaWaterResources Institute

References

[1] D F Heermann and K H Solomon ldquoEfficiency and unifor-mityrdquoDesign and operation of farm irrigation systems vol 5 pp108ndash119 2007

[2] N K Gontia and K N Tiwari ldquoEstimation of crop coefficientand evapotranspiration of wheat (Triticum aestivum) in anirrigation command using remote sensing and GISrdquo WaterResources Management vol 24 no 7 pp 1399ndash1414 2010

[3] H V Parmar and N K Gontia ldquoRemote sensing based vegeta-tion indices and crop coefficient relationship for estimation ofcrop evapotranspiration in Ozat-II canal commandrdquo Journal ofAgrometeorology vol 18 no 1 pp 137ndash139 2016

[4] E Adamala Y A Rajwade and Y V K Reddy ldquoEstimation ofwheat crop evapotranspiration using NDVI vegetation indexrdquoJournal of Applied and Natural Science vol 8 no 1 pp 159ndash1662016

[5] R G Allen L S Pereira D Raes and M Smith Cropevapotranspiration Guide-lines for computing crop requirementsIrrigation and Drainage Paper no 56 FAO Rome Italy 1998

[6] R G Allen A J Clemmens C M Burt K Solomon and TOrsquoHalloran ldquoPrediction accuracy for projectwide evapotranspi-ration using crop coefficients and reference evapotranspirationrdquoJournal of Irrigation andDrainage Engineering vol 131 no 1 pp24ndash36 2005

[7] M E Jensen and R G Allen Evaporation Evapotranspirationand Irrigation Water Requirements ASCE Manuals and reports


on Engineering Practice no 70 American Society of CivilEngineers Reston Virginia USA 2nd edition 2016

[8] W G M Bastiaanssen E J M Noordman H Pelgrum GDavids B P Thoreson and R G Allen ldquoSEBAL model withremotely sensed data to improve water-resources managementunder actual field conditionsrdquo Journal of Irrigation andDrainageEngineering vol 131 no 1 pp 85ndash93 2005

[9] R Allen M Tasumi and R Trezza ldquoSatellite-based energybalance for mapping evapotranspiration with internalized cali-bration (METRIC)mdashmodelrdquo Journal of Irrigation and DrainageEngineering vol 133 no 4 pp 380ndash394 2007

[10] J Kjaersgaard R Allen and A Irmak ldquoImproved methods forestimating monthly and growing season ET using METRICapplied to moderate resolution satellite imageryrdquo HydrologicalProcesses vol 25 no 26 pp 4028ndash4036 2011

[11] A Reyes-Gonzalez J Kjaersgaard T Trooien C Hay andL Ahiablame ldquoComparative Analysis of METRIC Model andAtmometer Methods for Estimating Actual Evapotranspira-tionrdquo International Journal of Agronomy vol 2017 pp 1ndash16 2017

[12] C M U Neale H Jayanthi and J L Wright ldquoIrrigation watermanagement using high resolution airborne remote sensingrdquoIrrigation and Drainage Systems vol 19 no 3-4 pp 321ndash3362005

[13] E P Glenn C M U Neale D J Hunsaker and P L NaglerldquoVegetation index-based crop coefficients to estimate evapo-transpiration by remote sensing in agricultural and naturalecosystemsrdquo Hydrological Processes vol 25 no 26 pp 4050ndash4062 2011

[14] J W Rouse R H Haas J A Schell and D W DeeringldquoMonitoring vegetation systems in the Great Plains with ERTSrdquoinThird ERTS symposium pp 309ndash317 NASAWashingtonDCUSA 10-14 December 1973

[15] E P Glenn P L Nagler and A R Huete ldquoVegetation IndexMethods for Estimating Evapotranspiration by Remote Sens-ingrdquo Surveys in Geophysics vol 31 no 6 pp 531ndash555 2010

[16] C Romero-Trigueros P A Nortes J J Alarcon et al ldquoEffectsof saline reclaimed waters and deficit irrigation on Citrus phys-iology assessed by UAV remote sensingrdquo Agricultural WaterManagement vol 183 pp 60ndash69 2017

[17] C Toureiro R Serralheiro S Shahidian and A Sousa ldquoIrri-gation management with remote sensing evaluating irrigationrequirement formaize underMediterranean climate conditionrdquoAgricultural Water Management vol 184 pp 211ndash220 2017

[18] H Lei and D Yang ldquoCombining the crop coefficient of winterwheat and summer maize with a remotely sensed vegetationindex for estimating evapotranspiration in the North Chinaplainrdquo Journal of Hydrologic Engineering vol 19 no 1 pp 243ndash251 2014

[19] E G Kullberg K C DeJonge and J L Chavez ldquoEvaluationof thermal remote sensing indices to estimate crop evapotran-spiration coefficientsrdquoAgriculturalWaterManagement vol 179pp 64ndash73 2017

[20] C M U Neale W C Bausch and D F Herman ldquoDevelopmentof reflectance-based crop coefficient for cornrdquo Transactions ofthe ASAE vol 32 no 6 pp 1891ndash1899 1989

[21] W C Bausch ldquoRemote sensing of crop coefficients for improv-ing the irrigation scheduling of cornrdquo Agricultural Water Man-agement vol 27 no 1 pp 55ndash68 1995

[22] J Garatuza-Payan A Tamayo C Watts and J RodriguezldquoEstimating large area wheat evapotranspiration from remotesensing datardquo in Proceedings of the IGARSS 2003 2003 IEEE

International Geoscience and Remote Sensing Symposium pp380ndash382 Toulouse France

[23] D J Hunsaker P J Pinter Jr E M Barnes and B A KimballldquoEstimating cotton evapotranspiration crop coefficients with amultispectral vegetation indexrdquo Irrigation Science vol 22 no 2pp 95ndash104 2003

[24] M Tasumi R G Allen R Trezza and J L Wright ldquoSatellite-based energy balance to assess within-population variance ofcrop coefficient curvesrdquo Journal of Irrigation and DrainageEngineering vol 131 no 1 pp 94ndash109 2005

[25] B Duchemin R Hadria S Erraki et al ldquoMonitoring wheatphenology and irrigation in Central Morocco On the use ofrelationships between evapotranspiration crops coefficientsleaf area index and remotely-sensed vegetation indicesrdquo Agri-cultural Water Management vol 79 no 1 pp 1ndash27 2006

[26] H Jayanthi C M U Neale and J L Wright ldquoDevelopmentand validation of canopy reflectance-based crop coefficient forpotatordquo Agricultural Water Management vol 88 no 1-3 pp235ndash246 2007

[27] T J Trout ldquoRemote sensing of canopy cover in horticulturalcropsrdquo HortScience vol 43 no 2 pp 333ndash337 2008

[28] M P Gonzalez-Dugo and L Mateos ldquoSpectral vegetationindices for benchmarking water productivity of irrigated cottonand sugarbeet cropsrdquo Agricultural Water Management vol 95no 1 pp 48ndash58 2008

[29] I Campos C M U Neale A Calera C Balbontın and JGonzalez-Piqueras ldquoAssessing satellite-based basal crop coeffi-cients for irrigated grapes (Vitis vinifera L)rdquoAgriculturalWaterManagement vol 98 no 1 pp 45ndash54 2010

[30] B Kamble A Kilic and K Hubbard ldquoEstimating crop coef-ficients using remote sensing-based vegetation indexrdquo RemoteSensing vol 5 no 4 pp 1588ndash1602 2013

[31] A Reyes-Gonzalez C Hay J Kjaersgaard and C M U NealeldquoUse of remote sensing to generate crop coefficient and estimateactual crop evapotranspirationrdquo inASABEAnnual internationalmeeting New Orleans LA USA 2015

[32] E Farg S M Arafat M S Abd El-Wahed and A M El-GindyldquoEstimation of Evapotranspiration ETc and Crop Coefficient KcofWheat in southNile Delta of Egypt Using integrated FAO-56approach and remote sensing datardquo Egyptian Journal of RemoteSensing and Space Science vol 15 no 1 pp 83ndash89 2012

[33] S Vanino G Pulighe P Nino C de Michele S F Bolognesiand G DrsquoUrso ldquoEstimation of evapotranspiration and cropcoefficients of tendone vineyards using multi-sensor remotesensing data in a mediterranean environmentrdquo Remote Sensingvol 7 no 11 pp 14708ndash14730 2015

[34] H Zhang R G Anderson and D Wang ldquoSatellite-basedcrop coefficient and regional water use estimates for Hawaiiansugarcanerdquo Field Crops Research vol 180 pp 143ndash154 2015

[35] SAGARPA ldquoResumen Sector Agropecuario en la RegionLagunerardquo Publicacion especial el Siglo de Torreon p 24 2016

[36] R P Cano andMM CMedina ldquoTecnologia de produccion denogal pecanerordquo Libro Tecnico vol no 3 p 220 2002

[37] G Levin G A Cruz D Garcia C Garces-Restrepo and SJohnson Performance of two transferredmodules in the laguneraregion water relations Research Report 23 vol 23 nternationalWater Management Institute Colombo Sri Lanka 1998

[38] W C Bausch ldquoSoil background effects on reflectance-basedcrop coefficients for cornrdquo Remote Sensing of Environment vol46 no 2 pp 213ndash222 1993


[39] A Bannari D Morin F Bonn and A R Huete ldquoA review ofvegetation indicesrdquo International Journal of Remote Sensing vol13 no 1-2 pp 95ndash120 1995

[40] A Reyes-Gonzalez U Figueroa D G Reta J I Sanchezand J G Martinez Estimacion de la evapotranspiracion actualutilizando sensors remotos y mediciones in situ En la ComarcaLagunera Mexico 2012

[41] ASCE-EWRI ldquoThe ASCE Standardized Reference Evapotran-spiration Equation Report of theASCE-EWRITaskCommitteeon Standardization of Reference Evapotranspirationrdquo TechRep ASCE Reston VA USA 2005

[42] T J Jackson D Chen M Cosh et al ldquoVegetation water contentmapping using Landsat data derived normalized differencewater index for corn and soybeansrdquo Remote Sensing of Environ-ment vol 92 no 4 pp 475ndash482 2004

[43] P-Y Chen G Fedosejevs M Tiscareno-Lopez and J GArnold ldquoAssessment of MODIS-EVI MODIS-NDVI andVEGETATION-NDVI composite data using agricultural mea-surements An example at corn fields in western MexicordquoEnvironmental Modeling amp Assessment vol 119 no 1-3 pp 69ndash82 2006

[44] W E Thomason S B Phillips and F D Raymond ldquoDefininguseful limits for spectral reflectance measures in cornrdquo Journalof Plant Nutrition vol 30 no 8 pp 1263ndash1277 2007

[45] R K Singh and A Irmak ldquoEstimation of crop coefficients usingsatellite remote sensingrdquo Journal of Irrigation and DrainageEngineering vol 135 no 5 pp 597ndash608 2009

[46] CHW de Souza EMercante J A Johann RAC Lamparelliand M A Uribe-Opazo ldquoMapping and discrimination ofsoya bean and corn crops using spectro-temporal profiles ofvegetation indicesrdquo International Journal of Remote Sensing vol36 no 7 pp 1809ndash1824 2015

[47] F Gao M C Anderson X Zhang et al ldquoToward mapping cropprogress at field scales through fusion of Landsat and MODISimageryrdquo Remote Sensing of Environment vol 188 pp 9ndash252017

[48] N Flood ldquoContinuity of reflectance data between landsat-7ETM+ and landsat-8 OLI for both top-of-atmosphere andsurface reflectance a study in the australian landscaperdquo RemoteSensing vol 6 no 9 pp 7952ndash7970 2014

[49] Y Ke J Im J Lee H Gong and Y Ryu ldquoCharacteristics ofLandsat 8 OLI-derived NDVI by comparison with multiplesatellite sensors and in-situ observationsrdquo Remote Sensing ofEnvironment vol 164 pp 298ndash313 2015

[50] D P Roy V Kovalskyy H K Zhang et al ldquoCharacterization ofLandsat-7 to Landsat-8 reflective wavelength and normalizeddifference vegetation index continuityrdquo Remote Sensing of Envi-ronment vol 185 pp 57ndash70 2016

[51] D P Roy M A Wulder T R Loveland et al ldquoLandsat-8 science and product vision for terrestrial global changeresearchrdquo Remote Sensing of Environment vol 145 pp 154ndash1722014

[52] C E Holden andC EWoodcock ldquoAn analysis of Landsat 7 andLandsat 8 underflight data and the implications for time seriesinvestigationsrdquo Remote Sensing of Environment vol 185 pp 16ndash36 2016

[53] J Rocha A Perdigao R Melo and C Henriques ldquoRemotesensing based crop coefficients for water management in agri-cultureChapter 8rdquo INTECH pp 167ndash192 2012

[54] A Reyes-Gonzalez T Trooien J Kjaersgaard C Hay and D GReta-Sanchez ldquoDevelopment of crop coefficients using remote

sensing-based vegetation index and growing degree daysrdquoin Proceedings of the ASABE Annual International MeetingOrlando Fla USA July 2016

[55] E B Rafn B Contor and D P Ames ldquoEvaluation of a methodfor estimating irrigated crop-evapotranspiration coefficientsfrom remotely sensed data in Idahordquo Journal of Irrigation andDrainage Engineering vol 134 no 6 pp 722ndash729 2008

[56] R G Steel and J H Torrie Principles and procedures of statistica biometrical approach McGraw-Hill NY USA 1980

[57] A Irmak I Ratcliffe P Ranade et al ldquoEstimation of landsurface evapotranspiration with a satellite remote sensing pro-cedurerdquo Great Plains Research vol 21 no 1 pp 73ndash88 2011

[58] L Rossato R C S Alvala N J Ferreira and J TomasellaldquoEvapotranspiration estimation in the Brazil using NDVI datardquoin Proceedings of the Remote Sensing for Agriculture Ecosystemsand Hydrology VII Belgium September 2005

[59] MMahour V Tolpekin A Stein and A Sharifi ldquoA comparisonof two downscaling procedures to increase the spatial resolutionof mapping actual evapotranspirationrdquo ISPRS Journal of Pho-togrammetry and Remote Sensing vol 126 pp 56ndash67 2017

[60] M P Gonzalez-Dugo C M U Neale L Mateos et al ldquoAcomparison of operational remote sensing-based models forestimating crop evapotranspirationrdquo Agricultural and ForestMeteorology vol 149 no 11 pp 1843ndash1853 2009

[61] R S Murray P L Nagler K Morino and E P Glenn ldquoAnempirical algorithm for estimating agricultural and riparianevapotranspiration using MODIS enhanced vegetation indexand ground measurements of ET II application to the lowerColorado river USrdquo Remote Sensing vol 1 no 4 pp 1125ndash11382009

[62] S C Zipper and S P Loheide ldquoUsing evapotranspiration toassess drought sensitivity on a subfield scale with HRMET ahigh resolution surface energy balancemodelrdquoAgricultural andForest Meteorology vol 197 pp 91ndash102 2014

[63] G B SenayM Friedrichs R K Singh andNMVelpuri ldquoEval-uating Landsat 8 evapotranspiration for water use mapping inthe Colorado River Basinrdquo Remote Sensing of Environment vol185 pp 171ndash185 2016

[64] USDA-NASS Farm and Ranch Irrigation Survey Table 22-Methods used in deciding when to irrigate USDA NationalAgricultural Statistics Service 2013

[65] J C Henggeler M D Dukes and B Q Mecham ldquoChapter13 Irrigation Schedulingrdquo in Irrigation L E Stetson and B QMecham Eds pp 491ndash564 Falls Church VAUSA 6th edition2011

Hindawiwwwhindawicom Volume 2018

Journal of

ChemistryArchaeaHindawiwwwhindawicom Volume 2018

Marine BiologyJournal of


BiodiversityInternational Journal of


EcologyInternational Journal of


Hindawiwwwhindawicom

Applied ampEnvironmentalSoil Science

Volume 2018

Forestry ResearchInternational Journal of



International Journal of

Geophysics

Environmental and Public Health

Journal of




Microbiology


Public Health Advances in

AgricultureAdvances in


Agronomy




MeteorologyAdvances in

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018Hindawiwwwhindawicom Volume 2018

ChemistryAdvances in

ScienticaHindawiwwwhindawicom Volume 2018


Geological ResearchJournal of

Analytical ChemistryInternational Journal of


Submit your manuscripts atwwwhindawicom


Torreoacuten Coah

0 5 10 20(Km)

Meacutexico

1000 km500 mi

26∘0 0

N

25∘45 0

N

25∘30 0

N

26∘0 0

N

25∘45 0

N

25∘30 0

N

103∘400W 103∘200W 103∘00W

103∘400W 103∘200W 103∘00W

(b)

(a)

Figure 1 Location of the study area at northern Mexico (a) The subset of the area of interest Landsat with false color composite (bands 43 2) the yellow triangles denote five locations where we selected the corn fields and the white star indicates weather station (b)

Remote sensing is a technology that can estimate ETcat regional and local scale in less time and with less cost[9 10] Remote sensing can also estimate crop coefficientsbased on spectral reflectance of vegetation indices (VIs) [512] The normalized difference vegetation index (NDVI) isthe most common VIs [13] NDVI takes into account thereflectance of red and near-infrared wavebands [14] wherered waveband is strongly absorbed by chlorophyll in leaves ofthe top layers while near-infrared wavebands is reflected bythemesophyll structure in leaves penetrating into deeper leaflayers in a healthy vegetation [15 16]High values ofNDVI arerelated to healthy and dense vegetation which presents highreflectance values in the NIR waveband and low reflectancevalues in the red waveband [17] Crop coefficients generatedfrom VIs determine ETc better than a tabulated119870c because itrepresents the actual crop growth conditions and capture thespatial variability among different fields [2 18 19]

Several studies have usedmultispectral vegetation indicesderived from remote sensing to estimate119870c values on agricul-tural crops including corn crop for example in [20ndash31] Cropcoefficients derived from remotely sensed vegetation indexalso have used to generate local and regional ETc maps [2 32ndash34] however in northern Mexico ETc maps using satelliteremote sensing-based vegetation index remain unexplored

The objectives of this study were to (1) calculate NDVIvalues for each corn field for each growing season (2)develop a simple linear regression model between NDVI

derived from satellite-based remote sensing and tabulated119870cobtained of alfalfa-based crop coefficient fromASCEManual70 (3) generate119870c maps using the linear regression equationobtained between NDVI and 119870c values and (4) create ETcmaps with high spatial resolution at regional and field scales

2 Materials and Methods

21 Study Area The study was carried out in northernMexico (Comarca Lagunera) during four growing seasonsComarca Lagunera had an average latitude of 25∘401015840N andlongitude of 103∘181015840Wand elevation of 1115m abovemean sealevel (Figure 1) In Comarca Lagunera forage crops (alfalfacorn sorghum and oat (planted in the winter season))occupiedmore than 75 of the total irrigated area [35] Silagecorn is the most important crop after alfalfa in this regionFive silage corn fields in each growing season were selectedfor NDVI calculations The corn fields were irrigated usingsurface irrigation systems The plant population density was78000 plants haminus1 Silage corn is typically planted from lateMarch to early April and chopped for silage from late July toearly August depending on the crop variety The corn fieldsselected ranged between 10 and 20 hectares in size The soiltexture for this region is clay loam soil The mean annualmaximum temperature is 28∘C minimum 13∘C and mean21∘C [36] The mean annual precipitation is 200mm whilethe annual potential evapotranspiration is 2000mm [37]




2013







2014















2016

















Weather data


NDVI

Inputs

Calculations

Output

Landsat image


Crop ET(４＝)











119888MapsDevelop-











L7L8

15 25 35 45 55 65 75 85 95 1055DAP (2013)

0010203040506070809

ND

VI


Field 4Field 5

L7 L7 L8L8

0010203040506070809

ND

VI

15 25 35 45 55 65 75 85 95 1055DAP (2014)


Field 4Field 5

L7L8

15 25 35 45 55 65 75 85 95 1055DAP (2016)

0010203040506070809

ND

VI


Field 4Field 5

15 25 35 45 55 65 75 85 95 1055DAP (2015)

0010203040506070809

ND

VI


Field 4Field 5









33 119870119888Maps and 119870




2014

02 04 06 08 10NDVI

0

02

04

06

08

1

Kc


Field 4Field 5

2013

02 04 06 08 10NDVI

0

02

04

06

08

1

Kc


Field 4Field 5

2016

02 04 06 08 10NDVI

0

02

04

06

08

1

Kc


Field 4Field 5

2015

02 04 06 08 10NDVI

0

02

04

06

08

1

Kc


Field 4Field 5

Kc = 13301＄６） + 00021R2 = 09875

Kc = 14556＄６） + 00618

R2 = 0998

Kc = 12234＄６） + 00242

R2 = 09586

Kc = 10968＄６） + 01054

R2 = 09563

11 line

11 line 11 line

11 line





ALL

R2 = 097

Kc = 12358＄６） + 00245

02 04 06 08 10NDVI

0

02

04

06

08

1

Kc







DAP 48



Kc

00 05 10





24 32 40 48 56 64 80 88 96 1048DAP (2014)

00

02

04

06

08

10

12

Kc




y = 09684x + 00302

R2 = 09588

02 04 06 08 10 1200Kc table

00

02

04

06

08

10

12

Kc

calc

ulat

ed

11 line



DAP 48



0 4 8４＝ (mmday)







DAP 48



Corn field

8

4

0

４＝ (mmday)








Water save

0

2

4

6

8ET

(mm

daＳminus1)

15 25 35 45 55 65 75 85 95 1055DAP (2014)

４Ｌ

４＝




4 Conclusions










Acknowledgments


References








































































Journal of










Volume 2018





Geophysics


Journal of




Microbiology





Agronomy


















2013







2014















2016

















Weather data


NDVI

Inputs

Calculations

Output

Landsat image


Crop ET(４＝)











119888MapsDevelop-











L7L8

15 25 35 45 55 65 75 85 95 1055DAP (2013)

0010203040506070809

ND

VI


Field 4Field 5

L7 L7 L8L8

0010203040506070809

ND

VI

15 25 35 45 55 65 75 85 95 1055DAP (2014)


Field 4Field 5

L7L8

15 25 35 45 55 65 75 85 95 1055DAP (2016)

0010203040506070809

ND

VI


Field 4Field 5

15 25 35 45 55 65 75 85 95 1055DAP (2015)

0010203040506070809

ND

VI


Field 4Field 5









33 119870119888Maps and 119870




2014

02 04 06 08 10NDVI

0

02

04

06

08

1

Kc


Field 4Field 5

2013

02 04 06 08 10NDVI

0

02

04

06

08

1

Kc


Field 4Field 5

2016

02 04 06 08 10NDVI

0

02

04

06

08

1

Kc


Field 4Field 5

2015

02 04 06 08 10NDVI

0

02

04

06

08

1

Kc


Field 4Field 5

Kc = 13301＄６） + 00021R2 = 09875

Kc = 14556＄６） + 00618

R2 = 0998

Kc = 12234＄６） + 00242

R2 = 09586

Kc = 10968＄６） + 01054

R2 = 09563

11 line

11 line 11 line

11 line





ALL

R2 = 097

Kc = 12358＄６） + 00245

02 04 06 08 10NDVI

0

02

04

06

08

1

Kc







DAP 48



Kc

00 05 10





24 32 40 48 56 64 80 88 96 1048DAP (2014)

00

02

04

06

08

10

12

Kc




y = 09684x + 00302

R2 = 09588

02 04 06 08 10 1200Kc table

00

02

04

06

08

10

12

Kc

calc

ulat

ed

11 line



DAP 48



0 4 8４＝ (mmday)







DAP 48



Corn field

8

4

0

４＝ (mmday)








Water save

0

2

4

6

8ET

(mm

daＳminus1)

15 25 35 45 55 65 75 85 95 1055DAP (2014)

４Ｌ

４＝




4 Conclusions










Acknowledgments


References








































































Journal of










Volume 2018





Geophysics


Journal of




Microbiology





Agronomy
















Weather data


NDVI

Inputs

Calculations

Output

Landsat image


Crop ET(４＝)











119888MapsDevelop-











L7L8

15 25 35 45 55 65 75 85 95 1055DAP (2013)

0010203040506070809

ND

VI


Field 4Field 5

L7 L7 L8L8

0010203040506070809

ND

VI

15 25 35 45 55 65 75 85 95 1055DAP (2014)


Field 4Field 5

L7L8

15 25 35 45 55 65 75 85 95 1055DAP (2016)

0010203040506070809

ND

VI


Field 4Field 5

15 25 35 45 55 65 75 85 95 1055DAP (2015)

0010203040506070809

ND

VI


Field 4Field 5









33 119870119888Maps and 119870




2014

02 04 06 08 10NDVI

0

02

04

06

08

1

Kc


Field 4Field 5

2013

02 04 06 08 10NDVI

0

02

04

06

08

1

Kc


Field 4Field 5

2016

02 04 06 08 10NDVI

0

02

04

06

08

1

Kc


Field 4Field 5

2015

02 04 06 08 10NDVI

0

02

04

06

08

1

Kc


Field 4Field 5

Kc = 13301＄６） + 00021R2 = 09875

Kc = 14556＄６） + 00618

R2 = 0998

Kc = 12234＄６） + 00242

R2 = 09586

Kc = 10968＄６） + 01054

R2 = 09563

11 line

11 line 11 line

11 line





ALL

R2 = 097

Kc = 12358＄６） + 00245

02 04 06 08 10NDVI

0

02

04

06

08

1

Kc







DAP 48



Kc

00 05 10





24 32 40 48 56 64 80 88 96 1048DAP (2014)

00

02

04

06

08

10

12

Kc




y = 09684x + 00302

R2 = 09588

02 04 06 08 10 1200Kc table

00

02

04

06

08

10

12

Kc

calc

ulat

ed

11 line



DAP 48



0 4 8４＝ (mmday)







DAP 48



Corn field

8

4

0

４＝ (mmday)








Water save

0

2

4

6

8ET

(mm

daＳminus1)

15 25 35 45 55 65 75 85 95 1055DAP (2014)

４Ｌ

４＝




4 Conclusions










Acknowledgments


References








































































Journal of










Volume 2018





Geophysics


Journal of




Microbiology





Agronomy
















L7L8

15 25 35 45 55 65 75 85 95 1055DAP (2013)

0010203040506070809

ND

VI


Field 4Field 5

L7 L7 L8L8

0010203040506070809

ND

VI

15 25 35 45 55 65 75 85 95 1055DAP (2014)


Field 4Field 5

L7L8

15 25 35 45 55 65 75 85 95 1055DAP (2016)

0010203040506070809

ND

VI


Field 4Field 5

15 25 35 45 55 65 75 85 95 1055DAP (2015)

0010203040506070809

ND

VI


Field 4Field 5









33 119870119888Maps and 119870




2014

02 04 06 08 10NDVI

0

02

04

06

08

1

Kc


Field 4Field 5

2013

02 04 06 08 10NDVI

0

02

04

06

08

1

Kc


Field 4Field 5

2016

02 04 06 08 10NDVI

0

02

04

06

08

1

Kc


Field 4Field 5

2015

02 04 06 08 10NDVI

0

02

04

06

08

1

Kc


Field 4Field 5

Kc = 13301＄６） + 00021R2 = 09875

Kc = 14556＄６） + 00618

R2 = 0998

Kc = 12234＄６） + 00242

R2 = 09586

Kc = 10968＄６） + 01054

R2 = 09563

11 line

11 line 11 line

11 line





ALL

R2 = 097

Kc = 12358＄６） + 00245

02 04 06 08 10NDVI

0

02

04

06

08

1

Kc







DAP 48



Kc

00 05 10





24 32 40 48 56 64 80 88 96 1048DAP (2014)

00

02

04

06

08

10

12

Kc




y = 09684x + 00302

R2 = 09588

02 04 06 08 10 1200Kc table

00

02

04

06

08

10

12

Kc

calc

ulat

ed

11 line



DAP 48



0 4 8４＝ (mmday)







DAP 48



Corn field

8

4

0

４＝ (mmday)








Water save

0

2

4

6

8ET

(mm

daＳminus1)

15 25 35 45 55 65 75 85 95 1055DAP (2014)

４Ｌ

４＝




4 Conclusions










Acknowledgments


References








































































Journal of










Volume 2018





Geophysics


Journal of




Microbiology





Agronomy
















2014

02 04 06 08 10NDVI

0

02

04

06

08

1

Kc


Field 4Field 5

2013

02 04 06 08 10NDVI

0

02

04

06

08

1

Kc


Field 4Field 5

2016

02 04 06 08 10NDVI

0

02

04

06

08

1

Kc


Field 4Field 5

2015

02 04 06 08 10NDVI

0

02

04

06

08

1

Kc


Field 4Field 5

Kc = 13301＄６） + 00021R2 = 09875

Kc = 14556＄６） + 00618

R2 = 0998

Kc = 12234＄６） + 00242

R2 = 09586

Kc = 10968＄６） + 01054

R2 = 09563

11 line

11 line 11 line

11 line





ALL

R2 = 097

Kc = 12358＄６） + 00245

02 04 06 08 10NDVI

0

02

04

06

08

1

Kc







DAP 48



Kc

00 05 10





24 32 40 48 56 64 80 88 96 1048DAP (2014)

00

02

04

06

08

10

12

Kc




y = 09684x + 00302

R2 = 09588

02 04 06 08 10 1200Kc table

00

02

04

06

08

10

12

Kc

calc

ulat

ed

11 line



DAP 48



0 4 8４＝ (mmday)







DAP 48



Corn field

8

4

0

４＝ (mmday)








Water save

0

2

4

6

8ET

(mm

daＳminus1)

15 25 35 45 55 65 75 85 95 1055DAP (2014)

４Ｌ

４＝




4 Conclusions










Acknowledgments


References








































































Journal of










Volume 2018





Geophysics


Journal of




Microbiology





Agronomy

















DAP 48



Kc

00 05 10





24 32 40 48 56 64 80 88 96 1048DAP (2014)

00

02

04

06

08

10

12

Kc




y = 09684x + 00302

R2 = 09588

02 04 06 08 10 1200Kc table

00

02

04

06

08

10

12

Kc

calc

ulat

ed

11 line



DAP 48



0 4 8４＝ (mmday)







DAP 48



Corn field

8

4

0

４＝ (mmday)








Water save

0

2

4

6

8ET

(mm

daＳminus1)

15 25 35 45 55 65 75 85 95 1055DAP (2014)

４Ｌ

４＝




4 Conclusions










Acknowledgments


References








































































Journal of










Volume 2018





Geophysics


Journal of




Microbiology





Agronomy
















y = 09684x + 00302

R2 = 09588

02 04 06 08 10 1200Kc table

00

02

04

06

08

10

12

Kc

calc

ulat

ed

11 line



DAP 48



0 4 8４＝ (mmday)







DAP 48



Corn field

8

4

0

４＝ (mmday)








Water save

0

2

4

6

8ET

(mm

daＳminus1)

15 25 35 45 55 65 75 85 95 1055DAP (2014)

４Ｌ

４＝




4 Conclusions










Acknowledgments


References








































































Journal of










Volume 2018





Geophysics


Journal of




Microbiology





Agronomy

















DAP 48



Corn field

8

4

0

４＝ (mmday)








Water save

0

2

4

6

8ET

(mm

daＳminus1)

15 25 35 45 55 65 75 85 95 1055DAP (2014)

４Ｌ

４＝




4 Conclusions










Acknowledgments


References








































































Journal of










Volume 2018





Geophysics


Journal of




Microbiology





Agronomy
















Water save

0

2

4

6

8ET

(mm

daＳminus1)

15 25 35 45 55 65 75 85 95 1055DAP (2014)

４Ｌ

４＝




4 Conclusions










Acknowledgments


References








































































Journal of










Volume 2018





Geophysics


Journal of




Microbiology





Agronomy















































































Journal of










Volume 2018





Geophysics


Journal of




Microbiology





Agronomy













































Journal of










Volume 2018





Geophysics


Journal of




Microbiology





Agronomy
















Journal of










Volume 2018





Geophysics


Journal of




Microbiology





Agronomy















Documents

Estimation of Crop Evapotranspiration Using Satellite ...downloads.hindawi.com/journals/amete/2018/4525021.pdf · Estimation of Crop Evapotranspiration Using Satellite Remote Sensing-Based