Paper Thirteen

header

Drip Irrigation Impacts on Evapotranspiration Rates in California’s San Joaquin Valley

Seventh International Conference on Irrigation and Drainage

Phoenix, Arizona
April 16-19, 2013

USCID
The U.S. society for irrigation and drainage professionals

Edited by

Brian T. Wahlin
WEST Consultants, Inc.
Susan S. Anderson
U.S. Committee on Irrigation and Drainage

Published by
U.S. Committee on Irrigation and Drainage
1616 Seventeenth Street, #483
Denver, CO 80202
Telephone: 303-628-5430
Fax: 303-628-5431
E-Mail: stephens@uscid.org
Internet: www.uscid.org

 

DRIP IRRIGATION IMPACTS ON EVAPOTRANSPIRATION RATES IN CALIFORNIA’S SAN JOAQUIN VALLEY

Bryan Thoreson1

Deepak Lal2

Byron Clark3

 

ABSTRACT

The acreage irrigated using drip irrigation continues to increase in California and the West. In addition to the increase on orchards and vineyard crops, more and more field crop acreage is coming under drip irrigation due to the increased production that growers are able to achieve with the increased distribution uniformity and improved water and nutrient management possible. When compared to surface irrigation methods, growers are able to increase production with the same, or often less, water applied. Research plot and analytical studies have shown increased evapotranspiration (ET) rates for drip irrigated field crops. Even with less water applied, increased ET rates equal more water consumed. Should the public be concerned about conversion to drip irrigation because water consumption may increase and supplies may be depleted, or should we be advocating conversion to drip/micro because we get more production per unit of water? This study addresses one technical aspect of this question by comparing ET rates over a large area of commercial production agriculture in California’s San Joaquin Valley using remotely sensed ET data together with field based land use information.

Using GIS-based crop and irrigation method data and SEBAL ET results, the evapotranspiration rates of populations of selected crop-irrigation method combinations will be compared and contrasted. Statistical tests will be applied to differences in the mean ET rates and within field ET rate variability between selected crop-irrigation method groups. The statistical comparisons will focus on the differences between ET rates of drip irrigation and other irrigation methods.

 

INTRODUCTION

The cropped area using drip and micro irrigation systems (drip/micro) in California has increased by about thirty-one percent from 1972 to 2001 (Orang, et al., 2008). A corresponding decrease in surface irrigation use has also occurred. These researchers also report an increase in the area planted to orchards and vineyards and a decrease in the area planted to field crops. However, much of the increase in drip/micro irrigation systems results from conversion of surface irrigated orchards and field crops to drip/micro. Understanding the effect of this continuing shift in irrigation methods on actual crop evapotranspiration (ETa) is important for water planning and management.

Numerous researchers have estimated the differences in seasonal ETa between crops irrigated with drip/micro systems and the same crops irrigated with surface irrigation systems. Burt, et al., (2002) estimated that orchard crop ETa in the San Joaquin Valley was six to 10 percent higher under drip/micro irrigation compared to surface or sprinkler irrigation. That study estimated the Transpiration (T) and Evaporation (E) components of ET using the FAO 56 (Allen, et al, 1998) dual crop coefficient method for various types of irrigation systems and irrigated areas of California. The study found total ET for furrow, sprinkle, and subsurface drip irrigation (SDI) to be nearly the same. However, the proportions of T and E comprising ET were different depending on the irrigation method, with SDI having the least evaporation of applied irrigation water (4% of seasonal ETa) and sprinkle irrigation having the most (8% of ETa). Hsiao et al. (2007) describe a systematic and quantitative approach to improve water use efficiency in agriculture.

They describe water use efficiency in agriculture in terms of a chain of efficiencies and discuss the reduction of E as one way to increase transpiration efficiency. In this discussion, they cite Bonachela et al. (2001), who notes that in tree crops, especially those with sparse canopy cover, E reductions achieved by localized irrigation (i.e., drip and micro) can be substantial.

Ward and Pulido-Velazquez (2008) analyze water conservation policies on an integrated, basin-scale linking biophysical, hydrologic, agronomic, economic, policy, and institutional dimensions. Based partly on large ETa increases ranging from 22 to 29 percent resulting from conversion from surface irrigation to drip irrigation, their results conclude that water conservation subsidies are unlikely to reduce water use. Their study demonstrates the importance of understanding changes in ETa resulting from conversion from surface to drip/micro irrigation. For example, if ETa does not increase substantially, and yield increases typically associated with drip/micro irrigation are due to a change in the partitioning of ET into E and T, the above mentioned study’s conclusions would not be valid.
Increases in ETa, or consumptive use of water, as the use of drip/micro increases throughout California could impact future water supplies. The objective of this paper is to further the knowledge and understanding of the changes in ETa resulting from changes in irrigation methods. This is approached by comparing seasonal ETa for a range of commercially produced crops irrigated with surface and drip/micro irrigation methods, and the statistical evaluation of the ETa differences.

Remotely sensed determination of actual ETa rates developed using a surface energy balance algorithm provide reliable seasonal estimates for individual fields under production conditions (Bastiaanssen et al., 2005). For this analysis, the SEBAL® (Surface Energy Balance Algorithm for Land) model, utilizing satellite-based remotely sensed data together with ground-based weather station data, was used to estimate ETa across a large area in the San Joaquin Valley of California. The remainder of this paper describes the ETa estimation method; the methodology for obtaining, comparing, and statistically evaluating the differences between field average seasonal ETa values for various crops under surface and drip/micro irrigation methods; and the results of the analysis and the conclusions reached.

ETa ESTIMATION METHOD

Conservation of energy at the Earth’s surface denotes a balance between net radiation reaching the Earth’s surface from the Sun and the sum of soil, sensible, and latent heat fluxes. Latent heat flux (energy per unit area per unit time) can be easily converted into ET flux (volume of water per unit area per unit time) based on the latent heat of vaporization and density of water. ET flux can be estimated as a closure term from estimates of the remaining fluxes (Equation 1).

where 𝜆 is the latent heat of vaporization of water, 𝝆w is the density of water, ETa is the actual ETa, Rn is the net radiation flux at the Earth’s surface, G is the soil heat flux, and H is the sensible heat flux.

Description of SEBAL

The SEBAL model applies radiative, aerodynamic, and energy balance physics in a series of 25 computational steps to estimate ETa from the energy balance. ETa is calculated at the pixel-scale using multispectral satellite imagery with a thermal band. The key input data consist of radiances in the visible, near infrared, and thermal infrared regions sensed by earth observing satellites; ground based weather data from agricultural or other weather stations; and land use data describing general vegetation types, when available. Knowledge of specific crop types is not needed to solve the energy balance. SEBAL is internally calibrated for each image to estimate sensible heat flux between the surface and the atmosphere, avoiding the need for absolute calibration of the surface temperature of each pixel. A detailed explanation of the algorithm is provided by Bastiaanssen et al. (1998).

SEBAL has been continually updated over time based on advances in surface energy balance science. These advances include both published and non-published refinements. The 2009 version of SEBAL used for this study includes the following updates from the originally published version of the model:

  • Topographic correction of extraterrestrial solar radiation based on actual surface slope and aspect,
  • Lapse rate correction of observed surface temperatures prior to calibration of sensible heat flux to normalize for elevation effects on surface temperature,
  • Use of spatially distributed weather surfaces from MeteoLook for improved representation of actual surface conditions within the image area,
  • Advection correction based on comparison of instantaneous and daily evaporative fractions estimated for a hypothetical grass reference surface assumed to be 0.12 m tall, having a surface resistance of 70 s and an albedo of 0.23 (Allen et. al., 1998) , which is used to compute a theoretical advection correction factor, which is then adjusted based on the actual instantaneous evaporative fraction for each pixel within the image,
  • Atmospheric correction and calibration (as needed) of albedo, and
  • Improved soil heat flux estimation based on a combination of LAI, and soil moisture.

Validation of SEBAL

SEBAL was developed through 20 years of research and validation. Validation is ongoing due to periodic refinements, sensitivity of model results to analyst judgments related to internal calibration, and interest in further quantifying the accuracy of the approach. The algorithm has been applied in 15 countries, including 11 states mostly in the western United States. Comparisons have been made to six different ET estimation methods for a variety of landscapes including irrigated pasture, sugar beets, riparian vegetation, playas, olives, rice, palm trees, cotton, wheat, sunflower, peaches, almonds, tomatoes, bare soil, grassland and forest.

Recent validations of SEBAL, summarized by Bastiaanssen et al. (2005), have shown seasonal ETa results generally fall within five percent of seasonal ETa determined from reliable ground-based measurements. ETa results from a 2002 SEBAL analysis for the Southern San Joaquin Valley were compared to lysimeter measurements on alfalfa and peaches (Cassel, 2006) and surface renewal measurements on tomatoes (Roberson, 2006). In each comparison, the difference between the SEBAL ETa and the ground-based estimates was five percent or less (Figure 1).

Additionally, SEBAL estimates of district-wide ETa for the Imperial Irrigation District were compared to an independent water balance (Thoreson et al. 2009). Annual ETa was calculated for the 1998 water year (October 1997 – September 1998) based on measured inflows and outflows. Total consumptive use from SEBAL was found to agree with the annual water balance within 1 percent.

Input Data

A combination of satellite, ground-based meteorological, topographic, and land cover classification data are utilized to quantify spatially distributed ETa. For this study, these datasets were obtained from the U.S. Geological Survey (USGS), CIMIS, and the U.S. Department of Agriculture (USDA). These data are described in greater detail in the following paragraphs.

Satellite Images

Seven Landsat 5TM and one Landsat 7 ETM multispectral images encompassing the period from late March to early November for Path 35/Row 42 were obtained from USGS for 2009 (Table 1). Cloud-free images were selected to achieve a temporal frequency of one image per month for each growing season.

Figure 1. Seasonal SEBAL ETa Results Compared to Lysimeter and Surface Renewal Results.

Table 1. SEBAL Datasets Used for 2009 Growing Season ET Analysis.

Region

Satellite Platform

Row/ Path

Thermal Resolution

Image Dates

Images

Southern San Joaquin Valley (2009 season)

Landsat 7 ETM

42/35

60 m

3/30/2009

1

Southern San Joaquin Valley (2009 season)

Landsat 5

42/35

120 m

4/23, 5/25, 6/26, 7/28, 8/29, 9/30, 11/1/2009

7

 

Meteorological Data

Measurements of incoming solar radiation (Rs), relative humidity (RH), air temperature (Ta) and wind speed (WS) were available as hourly averages for the time of image acquisition. Daily (average for the image date), and period (average for the days represented by an individual image) measurements were also available. Eight CIMIS stations falling within or on the edge of the study area were used to develop a spatially varying weather surface prior to the SEBAL image processing. Weather data were quality checked according to the guidelines specified in Appendix-D of the ASCE Task Committee Report on the Standardized Reference Evapotranspiration Equation (Allen et al., 2005).

Weather data were spatially interpolated using MeteoLook, a collection of algorithms developed to interpolate point weather observations based on the surface and terrain characteristics coupled with physically-based models (Voogt, M.P., 2006). Processes that influence surface weather conditions such as elevation, surface roughness, albedo, incoming radiation, land wetness, and distance to water bodies are represented in MeteoLook. This improved spatial distribution of weather data improves the ability to estimate surface conditions influencing the surface energy balance.

Landuse Data and Digital Elevation Model (DEM)

Information describing land use types within the southern San Joaquin Valley was obtained from the statewide land use data provided by the USDA National Agricultural Statistics Service (NASS) (available at datagateway.nrcs.usda.gov) for the year 2009. The NASS land use map utilized is a raster grid derived primarily from multiple satellite images obtained from RESOURCESAT – 1 (IRS – P6) across the 2009 growing season. The NASS land use map was resampled from its original spatial resolution of 56 meters to 30 meters to be consistent with other inputs for SEBAL. This land use data was used to estimate obstacle heights for different surfaces within the study area. These data have been developed by various means including analysis of satellite images along with inspection of aerial photographs and ground-surveys.

A DEM of one arc-second resolution (approximately 30 meters) was obtained from USGS and was used in SEBAL to incorporate the effects of the slope, aspect and elevation of the land into the energy balance.

METHODOLOGY

Two general data sources were utilized in this study. An existing SEBAL dataset provided ETa estimates at the pixel scale derived from Landsat imagery. Field boundaries, crops and irrigation methods were identified using cropping data from the California Department of Water Resources (DWR) land use survey for east Fresno County for 2009.

Field-scale average seasonal ETa in inches was calculated for field groups defined based on crop, irrigation method, and estimated fractional canopy cover in the east Fresno County area covered by the DWR land use survey (Figure 2). Field boundaries were buffered inward to identify areas in which ETa estimates were not affected by heat transfer processes occurring outside of the field (thermal contamination). Then, seasonal ETa for each pixel within each field of interest was averaged to estimate field-scale seasonal ETa. The data were filtered to remove fields with low NDVI during critical growth periods (suggesting very young crops or miss-classification) and to group fields based on estimates of fractional ground cover, so that comparisons could be made across fields of similar maturity, canopy structure, and/or cover crop presence. Finally, differences in average field-scale ETa were compared, and tested for statistical significance.

Figure 2. Location of Fields with Land Use Data.

 

Buffering of Field Boundaries

Field boundaries were buffered inward to identify cropped areas that were not affected by thermal contamination. Thermal contamination within field pixels occurs when lower resolution thermal pixels cross the field boundary and are affected by heat transfer processes outside of the field. For such pixels, the thermal radiance represents a weighted average of the radiance of the pixel area outside the field and the pixel area inside the field. To reduce the thermal contamination on field edges to acceptable levels, field boundaries were buffered 30 meters inward, and all fields with an area of less than five acres after buffering were dropped from the analysis.

Seasonal ETa and June 26, 2009 (mid-season) image NDVI values for each pixel within the buffered areas were extracted and imported to a Microsoft Access database for calculation of field averages.

Filtering of Field Data Based on NDVI

Field ETa values were filtered prior to the calculation of average seasonal ETa based on mid-season NDVI. Filters were applied primarily to separate fields of tree and vine crops into fractional cover classes based on NDVI presumably representing similar maturity, canopy structure, and/or cover crop presence.

Threshold NDVI values for the filters were estimated based on a relationship estimating fractional canopy cover from NDVI (Equation 2) after the form of Choudhury et al. (1994):

Table 2. Threshold NDVI Values Corresponding to Estimated 10 Percent Fractional Cover Increments from Equation 2.

 

RESULTS AND DISCUSSION

Eleven crops in the area bounded by the intersection of the 2009 East Fresno County DWR Land Use Survey and the SEBAL dataset had more than 1,000 acres irrigated by drip/micro irrigation methods (Table 3). Including processing tomatoes, 74 percent of which are irrigated with buried drip systems, these 12 crops accounted for just over 400,000 irrigated acres. Notably 44 percent of the area was irrigated with drip or micro irrigation. When buried drip, sprinklers and other methods are included, 49 percent of the area is irrigated with methods other than the traditional surface irrigation methods of furrow, border strip and basin.

As described previously, a 30 meter buffer was applied to the field boundaries, and pixels impacted by clouds and scan line gaps4 as well as all remaining fields less than five acres were removed from the analysis. This led to a field data set for consideration for the statistical analysis (Table 4). For the succeeding analyses, crop-irrigation method groups with the number of fields greater than 30 and the total acreage greater than 1,000 acres were included.

Table 3. Crops with Area Irrigated by Drip/Micro Greater Than 1,000 Acres. Source: East Fresno County DWR Land Use Survey, 2009

Crop

Total Acres

Drip/Micro Irrigation Methods, Acres

Surface Irrigation Methods, Acres

Buried Drip Irrigation Methods, Acres

Sprinklers and Other Irrigation Methods, Acres

Vineyards

202,167

88,250

106,170

215

7,532

Almonds

67,668

29,499

37,404

39

725

Oranges

35,303

29,122

5,802

0

379

Peaches and Nectarines

33,253

3,585

29,128

0

540

Pistachios

18,732

17,222

1,496

0

14

Plums

14,613

2,858

11,554

0

201

Tomatoes (processing)*

10,688

649

968

7,947

1,124

Mixed

6,397

1,293

4,244

583

278

Walnuts

6,183

1,152

4,966

0

65

Miscellaneous deciduous

5,917

2,526

3,023

0

367

Olives

1,329

1,119

210

0

0

Lemons

1,077

1,026

51

0

0

Totals

403,326

178,301

205,015

8,785

11,225

Percent

44%

51%

2%

3%

*Processing tomatoes are included as a crop of interest even though most of the acreage was irrigated with buried drip systems.

Table 4. Crop-Method Combination Areas Remaining After Buffering, Removing Pixels Impacted by Clouds and Scan Line Gaps (Landsat 7 ETM) and Removing Fields Less Than 5 Acres.

Crop-Method Group

No. of Fields

Acres

Vineyards, drip/micro

1,610

32,391

Vineyards, surface

1,861

31,771

Oranges, drip/micro

761

15,789

Peaches and Nectarines, surface

746

11,172

Almonds, drip/micro

370

8,731

Almonds, surface

423

7,641

Plums, surface

346

4,124

Oranges, surface

194

2,628

Peaches and Nectarines, drip/micro

113

1,447

Plums, drip/micro

88

1,034

Pistachios, surface

18

270

Pistachios, drip/micro

16

236

Tomatoes (processing), drip/micro

11

206

Tomatoes (processing), surface

1

20

Totals

6,558

117,461

The larger areas of drip/micro irrigated almond and orange orchards and vineyards in the lower fractional cover classes indicate a grower preference for drip/micro irrigation compared to surface irrigation (Table 5)

Table 5. Crop-Method Combination Areas and Number of Fields by Fractional Cover Range.

Fractional Cover Class

Crop-Method Group

NDVI Range

No. of Fields

Acres

No. of Fields

Acres

Almonds

< 0.3

< 0.395

112

1,948

96

1,309

Almonds

0.3 – 0.4

0.395 to 0.475

71

1,448

79

1,247

Almonds

0.4 – 0.5

0.475 to 0.55

89

2,195

102

1,852

Almonds

0.5 – 0.6

0.55 to 0.618

81

2,492

114

2,546

Almonds

> 0.6

>0.618

17

648

32

687

Oranges

< 0.3

< 0.395

307

6,109

75

954

Oranges

0.3 – 0.4

0.395 to 0.475

223

4,327

66

1,025

Oranges

0.4 – 0.5

0.475 to 0.55

190

4,522

43

535

Oranges

0.5 – 0.6

0.55 to 0.618

37

788

8

99

Oranges

> 0.6

>0.618

4

43

2

14

Peaches and Nectarines

< 0.3

< 0.395

19

196

66

837

Peaches and Nectarines

0.3 – 0.4

0.395 to 0.475

15

262

77

1,075

Peaches and Nectarines

0.4 – 0.5

0.475 to 0.55

27

333

197

2,934

Peaches and Nectarines

0.5 – 0.6

0.55 to 0.618

38

465

224

3,600

Peaches and Nectarines

> 0.6

>0.618

14

191

182

2,725

Plum

< 0.3

< 0.395

24

241

51

614

Plum

0.3 – 0.4

0.395 to 0.475

12

137

37

432

Plum

0.4 – 0.5

0.475 to 0.55

18

257

85

1,095

Plum

0.5 – 0.6

0.55 to 0.618

23

300

102

1,225

Plum

> 0.6

>0.618

11

99

71

758

Vineyards

< 0.3

< 0.395

866

17,630

842

13,635

Vineyards

0.3 – 0.4

0.395 to 0.475

359

7,585

581

11,138

Vineyards

0.4 – 0.5

0.475 to 0.55

128

2,815

219

3,865

Vineyards

0.5 – 0.6

0.55 to 0.618

82

1,565

123

1,887

Vineyards

> 0.6

>0.618

175

2,796

96

1,246

Average ETa for surface irrigated fields was greater than the average ETa for drip or micro irrigated for 21 of 25 crop-irrigation method-fractional cover groups. (Table 6). The 0.4 to 0.5 fractional cover group for oranges and plums had average ETa of drip or micro fields greater than the average ETa of surface irrigated fields. The two smallest fractional cover classes for peaches and nectarines were found to have an average ETa of drip or micro irrigated fields slightly larger than average ETa of the surface irrigated fields. With the notable exceptions of almonds and peaches and nectarines, the average ETa of the surface irrigated fields exceeded average ETa of drip/micro irrigated fields by the greatest amount in the smaller fractional cover groups. This is likely due to greater evaporation from surface irrigation of the younger trees. For all groups, the ETa difference between surface and drip or micro irrigated fields was less than the standard deviation.

Table 6. Crop-Method Combination Average ETa and Standard Deviation by Fractional Cover Range.

Drip Micro Irrigation Method Surface Irrigation Method Surface – Drop
Crop-Method Group Fractional Cover Class NDVI Range Avg. ETa, in StdDev, in Avg. ETa, in StdDev, in Avg. ETa, in

Almonds

<0.3

< 0.395

16.6

4.2

18.2

4.2

1.6

Almonds

0.3 – 0.4

0.395 to 0.475

34.0

4.7

35.7

4.7

1.8

Almonds

0.4 – 0.5

0.475 to 0.55

41.9

5.1

43.7

5.3

1.7

Almonds

0.5 – 0.6

0.55 to 0.618

47.6

5.5

49.6

4.9

2.0

Almonds

>0.6

>0.618

54.0

4.4

55.9

4.3

1.9

Oranges

<0.3

< 0.395

16.5

4.1

19.6

3.8

3.1

Oranges

0.3 – 0.4

0.395 to 0.475

32.4

4.7

34.8

4.5

2.4

Oranges

0.4 – 0.5

0.475 to 0.55

38.3

4.6

38.1

4.7

-0.2

Oranges

0.5 – 0.6

0.55 to 0.618

42.4

4.2

44.1

4.8

1.6

Oranges

>0.6

>0.618

42.9

3.7

45.0

4.0

2.1

Peaches and Nectarines

<0.3

< 0.395

22.1

5.0

19.6

4.9

-2.5

Peaches and Nectarines

0.3 – 0.4

0.395 to 0.475

31.8

6.2

31.5

5.5

-0.3

Peaches and Nectarines

0.4 – 0.5

0.475 to 0.55

35.4

5.5

38.8

4.7

3.3

Peaches and Nectarines

0.5 – 0.6

0.55 to 0.618

39.6

5.2

43.6

4.1

4.0

Peaches and Nectarines

>0.6

>0.618

45.5

4.7

48.8

3.9

3.3

Plum

<0.3

< 0.395

16.1

4.8

18.1

5.1

2.1

Plum

0.3 – 0.4

0.395 to 0.475

27.5

5.7

31.0

4.8

3.6

Plum

0.4 – 0.5

0.475 to 0.55

37.9

4.2

37.2

4.6

-0.7

Plum

0.5 – 0.6

0.55 to 0.618

42.2

3.8

43.0

4.6

0.8

Plum

>0.6

>0.618

47.7

5.2

48.5

3.6

0.8

Vineyards

<0.3

< 0.395

13.6

3.6

17.0

4.3

3.4

Vineyards

0.3 – 0.4

0.395 to 0.475

21.1

3.6

25.0

4.4

3.9

Vineyards

0.4 – 0.5

0.475 to 0.55

29.2

4.1

30.9

3.9

1.6

Vineyards

0.5 – 0.6

0.55 to 0.618

35.6

3.7

35.9

3.3

0.3

Vineyards

>0.6

>0.618

41.0

3.1

42.4

3.0

1.4

Statistical hypotheses were tested regarding the differences of the means of each group. Walpole and Myers (1978) define a statistical hypothesis as “an assumption or statement which may or may not be true, concerning one or more populations.” For the fifty populations considered in this study, a statistical hypothesis that the mean ETa of surface irrigated fields was equal to the mean ETa of drip/micro irrigated fields for each crop fractional cover category was formulated and tested as to whether it should be rejected at the 􏰂 = 0.05 level of confidence. This hypothesis is called the null hypothesis (H0). The rejection of this hypothesis leads to the acceptance of the alternative hypothesis (H1) that the means are not equal. This null hypothesis was rejected for four of the five crop- method-fractional cover groups for almonds and vineyards, the two crops with the greatest area and number of fields (Table 7). Overall all 25 crop-method-fractional cover groups, the null hypothesis was rejected for 13 groups accounting for 98,411 acres and accepted for 12 groups accounting for 18,318 acres. The null hypothesis that the means were equal was not rejected for all five groups of plums, the crop with the smallest area in the study.

Table 7. Crop-Method Combination Test Statistic and Statistical Hypothesis Test Results.

Crop-Method Group

Fractional Cover Class

NDVI Range

Surface – Drip Avg. ETa, in

Z test Statisitic

Z0.05/2

Reject Null Hypothesis (Alpha = 0.05)

Almonds

< 0.3

< 0.395

1.6

2.732

1.96

Yes

Almonds

0.3 – 0.4

0.395 to 0.475

1.8

2.283

1.96

Yes

Almonds

0.4 – 0.5

0.475 to 0.55

1.7

2.288

1.96

Yes

Almonds

0.5 – 0.6

0.55 to 0.618

2.0

2.594

1.96

Yes

Almonds

> 0.6

>0.618

1.9

1.428

1.96

No

Oranges

< 0.3

< 0.395

3.1

6.215

1.96

Yes

Oranges

0.3 – 0.4

0.395 to 0.475

2.4

3.840

1.96

Yes

Oranges

0.4 – 0.5

0.475 to 0.55

-0.2

-0.269

-1.96

No

Oranges

0.5 – 0.6

0.55 to 0.618

1.6

0.898

1.96

No

Oranges

> 0.6

>0.618

2.1

0.618

1.96

No

Peaches and Nectarines

< 0.3

< 0.395

-2.5

-1.950

-1.96

No

Peaches and Nectarines

0.3 – 0.4

0.395 to 0.475

-0.3

-0.178

-1.96

No

Peaches and Nectarines

0.4 – 0.5

0.475 to 0.55

3.3

3.011

1.96

Yes

Peaches and Nectarines

0.5 – 0.6

0.55 to 0.618

4.0

4.510

1.96

Yes

Peaches and Nectarines

> 0.6

>0.618

3.3

2.562

1.96

Yes

Plum

< 0.3

< 0.395

2.1

1.697

1.96

No

Plum

0.3 – 0.4

0.395 to 0.475

3.6

1.946

1.96

No

Plum

0.4 – 0.5

0.475 to 0.55

-0.7

-0.588

-1.96

No

Plum

0.5 – 0.6

0.55 to 0.618

0.8

0.918

1.96

No

Plum

> 0.6

>0.618

0.8

0.484

1.96

No

Vineyards

< 0.3

< 0.395

3.4

17.382

1.96

Yes

Vineyards

0.3 – 0.4

0.395 to 0.475

3.9

14.802

1.96

Yes

Vineyards

0.4 – 0.5

0.475 to 0.55

1.6

3.590

1.96

Yes

Vineyards

0.5 – 0.6

0.55 to 0.618

0.3

0.612

1.96

No

Vineyards

> 0.6

>0.618

1.4

3.709

1.96

Yes

These results contradict the increased ETa attributed to drip/micro irrigation reported by other investigators (Burt, 2002 and Ward, 2008). Four possible reasons for this are:

  1. The more precise irrigation management possible on drip/micro irrigated fields provides increased opportunities to practice irrigation management strategies that result in lower ETa.
    1. Vineyards on drip/micro are more likely to be deficit irrigated resulting in reduced ETa,
    2. Almonds irrigated with drip/micro systems are more likely to be managed utilizing Regulated Deficit Irrigation (RDI) (Goldhammer, 2005) practices with irrigation restricted during certain growth periods leading to reduced ETa
  2. A reduction in E resulting from conversion to drip or micro, without an offsetting increase in T (T may increase, but not as much as E decreases)
  3. Differences in the practice of using cover crops in orchards depending on the use of drip/micro or surface irrigation methods.
  4. If surface systems tend to occur in surface water areas with relatively abundant supplies and drip/micro tends to occur in GW areas or water short areas in general, it is possible that differences in water supply source explain the outcome that drip ETa is not greater than surface ETa.

CONCLUSION

Mean ETa from production fields irrigated with surface irrigation methods was found to be greater than mean ETa from fields irrigated with drip/micro for 13 of 25 crop- irrigation method-fractional cover groups. This result is surprising given the conventional wisdom supported by findings of some researchers that ETa from drip/micro irrigated fields is more than that from surface irrigation. To obtain the increased yield reported from the drip/micro irrigated fields, it is likely that the partitioning of E and T has changed with the volume of E from drip/micro irrigated fields being less than the E from surface irrigated fields. Additional research into seasonal ETa for drip/micro and surface irrigated fields is necessary to confirm or refute this preliminary conclusion. However, if proven to hold true by additional research, this indicates that conversion to drip/micro irrigation on orchard crops in the San Joaquin Valley of California is not substantially increasing the consumptive use of water. Thus, installation of drip/micro irrigation may, in many cases, increase transpiration efficiency by reducing the volume of evaporation without increasing overall ETa. Other changes resulting from installation of drip/micro irrigation systems may include reduction of deep percolation. Future studies are required to evaluate differences in ETa from surface irrigated and drip/micro irrigated field crops such as tomatoes and alfalfa.

REFERENCES

Allen, R.G., Pereira, L.S., Raes, D., and Smith, M. (1998). Crop Evapotranspiration. Irrigation and Drainage Paper No. 56. Food and Agriculture Organization of the United Nations. Rome, Italy.

Allen, R.A., Walter, I.A., Elliot, R., Howell, T., Itenfisu, D. and Jensen, M. (2005). The ASCE Standardized Reference Evapotranspiration Equation. ASCE. Reston, VA.

Bastiaanssen, W.G.M., Menenti, M., Feddes, R.A., and Holtslag, A.A.M. (1998). A Remote Sensing Surface Energy Balance Algorithm for Land (SEBAL): 1. Formulation. Journal of Hydrology, 212-213: 198-212.

Bastiaanssen, W.G.M., Noordman, E.J.M., Pelgrum, H., Davids, G., B.P. Thoreson and Allen, R.G. (2005). SEBAL Model with Remotely Sensed Data to Improve Water Resources Management under Actual Field Conditions. Journal of Irrigation and Drainage Engineering. 131(1), 85-93.

Bonachela, S.F, Orgaz, F., Villalobos, F.J, and Fereres, E. (2001) Soil evaporation from drip-irrigated olive orchards. Irrig Sci. 20:65–71. Cited in Hsiao, et al. 2007.

Burt, C. M., Mutziger, A., Howes, D. J., and Solomon, K. H. (2002). “Evaporation from irrigated land in California.” Rep. R02-00I. Irrigation Training and Research Center, California Polytechnic State Univ., San Luis Obispo, Calif. (http://www.itrc.orglreports/reportsindex.html).

Cassel, F. (2006). Remote Sensing of Evapotranspiration for Verification of Regulated Deficit Irrigation. USBR final project report. pp. 58.

Choudhury, B.J., Ahmed, N.U., Idso, S.B., Reginato, R.J., and Daughtry, C.S.T. (1994). Relations Between Evaporation Coefficients and Vegetation Indices Studied by Model Simulations. Remote Sensing of Environment. 50, 1-17.

Goldhammer, D.A. and Fereres, E . (2005). The Promise of Regulated Deficit Irrigation in California’s Orchards and Vineyards. California Water Plan, 2005. California Department of Water Resources, Sacramento, CA.

Hsiao, T. C., Steduto, P., and Fereres, E. (2007). A systematic and quantitative approach to improve water use efficiency in agriculture. Irrig Sci. 25:165–188

Orang, M.N., Matyac, J.S., and Snyder, R.L. (2008). “Survey of irrigation methods in California in 2001.” Journal of Irrigation and Drainage Engineering, 134 (1), 96-100.

Roberson, M. (2006). Personal communication.

Thoreson, B., Clark, B., Soppe, R., Keller, A., Bastiaanssen, W., and Eckhardt. J. (2009). Comparison of Evapotranspiration Estimates from Remote Sensing (SEBAL), Water Balance, and Crop Coefficient Approaches. In Great Rivers. Proceedings from an EWRI Congress, Kansas City, MO.

Voogt, M.P. (2006). Meteolook, a Physically Based Regional Distribution Model for Measured Meteorological Variables. MSc Thesis, TU Delft, the Netherlands.

Ward, F.A. and Pulico-Velazquez, M. (2008). Water Conservation in Irrigation Can Increase Water Use. Proceedings National Academy of Sciences. November 25, 2008 Vol. 105 No. 47, 18215-18220.

Walpole, R.E. and Myers, R.H. (1978). Probability and Statistics for Engineers and Scientists. Macmillan Publishing Co., Inc. New York, NY.

  1. Corresponding author, Davids Engineering, Inc., 1772 Picasso Avenue, Suite A, Davis, CA 95618, www.davidsengineering.com, bryan@davidsengineering.com.
  2. SEBAL North America, Inc., 1772 Picasso Avenue, Suite E, Davis, CA 95618, www.sebal.us, deepak@sebal.us.
  3. Davids Engineering, Inc., 1772 Picasso Avenue, Suite A, Davis, CA 95618, www.davidsengineering.com, byron@davidsengineering.com.
  4. Gaps present in the Landsat 7 image acquired March 30th result from a malfunction of the Scan Line Corrector (SLC) that occurred in May 2003 in the ETM+ imaging sensor onboard the satellite. Due to the gaps, approximately 22% of the data is missing from a typical Landsat 7 image. These gaps vary in width from one pixel or less in the center of the image to 14 pixels towards the edges of the image. Despite the gaps in the data, the Landsat 7 image was used in the analysis to provide adequate temporal coverage.