• 
• Japanese
• 
• IETC's Focal Area
• Disaster Management
• Projects
• Publications
• News
• Waste Management
• Projects
• Publications
• News
• Water and Sanitation
• Projects
• Publications
• News
• Resources
• Publications
• About IETC
• Background and History
• Director
• Activity Reports
• Supporting Foundations
• Procurement
• Contact Us

Newsletter

<Forum on the Caspian, Aral and Dead Seas-Perspective
of Water Environmental Management and Politics>

<Symposium on the Aral Sea and The Surrounding Region
-Irrigated Agriculture and the Environment>

Estimation of Evapotranspiration in both Irrigated and Non-Irrigated Lands in the Arid District of Central Asia

Kenichirou Kosugi, Kitano Nakajima, (Kyoto University, Fac. of Agriculture),
Yoshihiro Fukushima, (Nagoya University, Institute for Hydrospheric-Atmospheric Sciences)

Estimating evapotranspiration is important for the appropriate management of water resources in arid districts. Atmospheric variables and energy balances were measured at both irrigated and non-irrigated lands along the Ili river in Kazakhstan. With the aid of measured observations, the amounts of evapotranspiration were estimated. In the summer season, evapotranspiration was about 140 mm/month at all of the paddy field, barely field and alfalfa field sites of an irrigated farm. Evapotranspiration at non-irrigated land of Haloxylon (one of the typical coniferous trees in this region) was 25 mm/month.

I.Introduction

In the central Asian arid districts, many irrigated farms have been sited along the big river Amudarya, Syrdarya and Ili. This has brought large changes in land surface vegetation, and the amount of evapotranspiration has increased. Estimating the evapotranspiration of the various kinds of land surfaces is very important for appropriate management of water resources, especially in arid districts.

In order to estimate evapotranspiration, the process of energy exchange at the land surface should be analyzed. In the summer of 1992 and 1993, atomospheric variables and energy balances were measured at both irrigated and non-irrigated lands along the Ili river in the southern Balkhash region in Kazakhstan. Analyzing these measured observations, the amounts of evapotranspiration of the various kinds of land surfaces are estimated in this study.

II. Study site and methods

Measurements were made of a non-vegetation site, a paddy field, a barley field and an alfalfa of an irrigated farm along the Ili river in the summers of 1992 and 1993. Figure 1 shows the location of the irrigated farm. In this region, yearly precipitation was 114.3 mm and the yearly average temperature was about 9.1° C in 1991 (figure 2). The irrigated area of the farm 5,172 ha and about 40% of this area was used as paddy field.

The amount of water planned to be used for irrigation was about 1.5 x 10⁸ m³in 1993 (Watanabe, 1994). Figure 3 shows the soil moisture content profile at each experimental site. Ground water levels were about 80 and 40 cm from the soil surface at the barley field and the alfalfa field, respectively.

Measurements were also made at a shrublands and woodlands of Haloxylon in the summer of 1993. Haloxylon is one of the typical coniferous trees in this region. The shrubland was about 10 km from the irrigated farm situated and in a very dry condition. The woodland was in a wetter condition than the shrubland because it was situated in the irrigated farm (figure 3). Maximal vegetation height and the leaf area index (LAI) of each experimental site are summarized in table 1. The LAI of the woodland was about three times that of the shrubland.

Diurnal changes of air temperature (T), vapor pressure of air (e), wind velocity (U), solar radiation (Rs), long wave radiation (Rl), reflectivity (a), net radiation (Rn) and soil heat flux (G) were all observed. Sensible heat flux (H) was also measured the eddy correlation method using one-dimensional ultrasonic anemometer-thermometers at each site. Latent heat flux (lE) was computed using the energy balance equation.

III. Results and discussion

1.Diurnal changes of atmospheric variables

Figure 4 shows the diurnal changes of solar radiation (Rs) and air temperature (T), and figure 5 shows the diurnal changes of air saturation deficit (D) and wind velocity (U). One can see measurements were made under fine-weather conditions except at the barley field on July 1st, 1992 and at the shrubland of Haloxylon on August 10 in 1993. The weather was quite fine, especially on August 16 and 21 in 1993, and daily accumulated solar radiation was more than 27 MJ/m².
At the no-vegetation land, T and D increased to more than 40 °C and 60 hPa, respectively, in the afternoon. Values of T were smaller than 20 °C and values of D were smaller than 10 hPa all day at the paddy field. Values of T and D at the barley field were larger than those at the paddy field but were much smaller than those at the no-vegetation land. The maximal values of T and D at the alfalfa field were about 26 °C and 21 hPa. These are about the same as those at the barley field on July 1. At the shrubland of Haloxylon on the cloudy day (August 10), maximal value of T was about 29°C and the maximal value of D was about 28 hPa. On the fine days (August 13 and 16), T and D increased to more than 35°C and 45 hPa, respectively, both at the shrubland and at the woodland of Haloxylon.
Generally, wind velocity (U) was larger in the afternoon than in the morning. Values of U were large on July 2 in 1992 and August 10 in 1993, and small on July 1 in 1992 and August 21 in 1993.

2. Energy balances

The energy exchange process at the land surface is expressed by the following energy balance equation;

Equation (1) says that net radiation (Rn) is divided into soil heat flux (G), sensible heat flux (H), and latent heat flux (lE) at land surface. Sensible heat is used to increase air temperature and latent heat is used for vaporization of water. In equation (1) l expresses latent heat of water of unit weight and E expresses the amount of evapotranspiration.
Figure 6 shows the diurnal changes of Rn, G, H and lE at each experimental site. Daily accumulated values of Rs, Rn, G, H, lE, and E are summarized in table 2.
At the no-vegetation land, most of Rn was used as H, and lE was almost zero all day (figure 6a). Daily evapotranspiration was only 0.2 mm (table 2). The ratio of H to lE is called Bowen ratio and is used as an index to characterize the energy exchange process at land surface. Bowen ratio is small at the land surface where much energy is used for evaportranspiration. The ratio was also shown in table 2. One can see from the table that daily average Bowen ratio was quite large at the no-vegetation land. The diurnal changes of H and lE at the paddy field (figure 6b) contrast with those at the no-vegetation land. Most of Rn was consumed as lE, and H was smaller than 100W/m² all day. Daily evapotranspiration was 3.7 mm and daily average Bowen ratio was 0.14.
Figure 6c shows that H was less than 100 W/m² at the barley field on the cloudy day (July 1). On the fine day (July 2), H increased to about 200 W/m² in the early afternoon (figure 6d). Maximal value of lE was about 300 W/m² and daily evapotranspiration was about 3 mm both on the cloudy day and the fine day. Daily average Bowen ratio was quite small (equal to 0.06) on the cloudy day and was equal to 0.46 on the fine day.

Table 1. Maximal vegetation height and LAI of each experimental site

* from Kobayashi and Okitu, 1994

Figure 1: Location of the irrigated farm

Figure 2: Seasonal changes of temperature & precipitation

Figure 3: Soil moisture content profile

Figure 4: Diurnal changes of Rs and T

Figure 5: Diurnal changes of D and U

Figure 6: Diurnal changes of Rn, G, H and lE

Figure 4: Diurnal changes of Rs and T (continuation)

Figure 5: Diurnal changes of D and U (continuation)

Figure 6: Diurnal changes of Rn, G, H and lE (continuation)

Table 2. Daily accumulated values of Rs, Rn, G, H, lE, and E, and daily average ratio H/lE and lE/Rs, and estimated monthly evapotranspiration

* avereage value for two days
** estimated evapotranspiration for the one month period from August 8 to September 7 in 1993

More than 90% of Rn was consumed as lE at the alfalfa field (figure 6e). One can see that the changes of lE in the daytime was quite similar to the change of Rs shown in figure 4. Daily evapotranspiration was 5.1 mm. The daily average Bowen ratio at the alfalfa field (0.13) was small as that at the paddy field.

Figure 6f and 6g show that more than 80% of Rn was used as H at the shrubland of Haloxylon both on the cloudy day (August 10) and the fine day (August 13). On the cloudy day, lE was about 50 W/m² in the day time. On the fine day, lE showed a steep increase in the early morning followed by a gradual decline and became minus in the afternoon. These minus lE values may be attributable to differences in the sampling areas for H by the eddy correlaton method, and Rn by the net radio meter (Kelliher et al., 1994). The amounts of Daily evapotranspiration were 0.8 and 0.5 mm, and daily average Bowen ratio were 2.3 and 7.3 on the cloudy day and the fine day, respectively.

At the woodland, most of Rn was used as lE in the early morning (figure 6h). The value of lE was about 230 W/m² at 10:30, then gradually declined. Most of Rn was used as H in the late evening. Daily evapotranspiration (2.6 mm) was much larger than that at the shrubland. Daily average Bowen ratio was1.3.

The daily average ratio of lE to solar radiation (Rs) was also computed for each experimental site (table 2). For the period of one month from August 8 to September 7 in 1993, the accumulated solar radiation measured at the irrigated farm was728 MJ/m². Using this accumulated solar radiation and the ratio of lE to Rs, we estimated monthly evapotranspiration. The results are shown in table 2. Monthly evapotranspiration at the no-vegetation land was estimated to be only 6 mm. At the paddy field and the alfalfa field sites, estimated evapotranspiration was 140 and 136 mm/month, respectively, both of which were about the same as the estimated evapotranspiration at the paddy field (143 mm/month). A large amount of evapotranspiration at the barley field and the alfalfa field sites may be attributable to the high ground water level and large soil moisture content at these fields (see figure 3). Estimated monthly evapotranspiration was 70 mm at the woodland of Haloxylon, which estimated evapotranpiration was 25 mm/month at the shrubland of Haloxylon. As already shown in figure 3, he woodland was in a wetter condition than the shrubland as it was affected by the irrigation. This may be the reason that evapotranspiration at the wood land was about three times greater than that at the shrubland.

3. Analysis of canopy resistance and canopy conductance

In order to analyze the plant -atmospheric interaction, modeling canopy conductance (gc) or canopy resistance (rc) is very important. Assuming the whole vegetation canopy as 'a big leaf', sensible heat flux on the vegetation is regulated by the stomata of this big leaf. Canopy conductance (gc) is a parameter to characterize this regulation. Canopy resistance (rc) is the reciprocal of canopy conductance.

The values of gc or rc depend on both vegetation and atmospheric variables such as air temperature, air saturation deficit and solar radiation. Using the measured observations at the alfalfa field, and the shrubland and the woodland of Haloxylon, characteristics of gc or rc at both irrigated and non-irrigated lands were examined.

Aero dynamic resistance (ra) and canopy resistance (rc) were computed using the following bulk equations (Kosugi and Fukushima, 1994);

Figure 7 shows the diurnal changes of rc. Values of rc on the alfalfa field were equal to zero or quite small before 12:00 because of the evaporation of dew. They were about 50 sec/m from 12:00 till 18:00. The values of rc on the shrubland and on the woodland gradually increased with time. They were much greater than rc values at the alfalfa field. Values of rc at the shrubland were about twice greater than those at the woodland.

Values of gc at the alfalfa field, the shrubland and the woodland were computed using the rc values by the following equation;

Figure 8 and 9 show the relationship between gc and photosynthetically active radiation (PAR) and the relationship between gc and air saturation deficit (D), respectively. Here, values of PAR were derived from observed Rs values. Values of gc at the alfalfa field increased as PAR increased, while no apparent correlation was found between gc and D. Therefore gc mainly depends on PAR at the alfalfa field. Maximum canopy conductance was about 22 mm/sec. At the shrubland and the woodland of Haloxylon, gc was greater in the morning than in the afternoon under the same PAR condition. There were clear negative correlations between gc and D. These facts show that gc values depended mainly on D at the shrubland and the woodland. Values of maximum canopy conductance were estimated at about 5 and 10 mm/sec and the shrubland and the woodland, respectively.

Figure 7: Diurnal changes of rc

Figure 8: Relationship between gc and PAR

Figure 9: Relationship between gc and D

Acknowledgements:

This research was partly supported by a grant from the Toyota Foundation for Scientific Research. We express our gratitude to Dr.N.Ishida, Dr.T. Kobayashi, Dr.N.Ohte, N.Nakamura and Y.Kosugi for their support and help.

References :

Kelliher, F. M.,. Y. Hollinger, E-D. Schulze, N. N. Vygodskaya, J. N. Byers, J. E. Hunt, and T. M. Mcseveny, Evaporation from an eastern Siberian larch forest, Proceeding of International Symposium on Forest Hydrol., Tokyo Japan, 123-130,1994.

Kobayashi, T. and J. Okitu, Analysis of water dynamics and photosynthesis of Haloxyon and some other plants growing near an irrigated farm, Research Report of Jpn. Res. Assoc. of Kazakhstan, 57-73, 1994. (in Japanese - The title is a tentative translation from the original Japanese title by the authors of this abstract.)

Kosugi, K. and Y. Fukushima, Estimation of evapotranspiration at alfalfa field on irrigated land in arid district, J. Japan Soc. Hydrol. and Water Resour., 7(5), 420-429,1994. (in Japanese with English summary)

Watanabe, T., Irrigation system and amount of irrigation water used in Bereke national farm, Research Report of Jpn. Res.Assoc. of Kazakhstan, 57-73, 1994. (in Japanese- The title is a tentative translation from the original Japanese title by the authors of this abstract.)

Table of Contents

• Major Projects
• 
• DEBRI Project
• 
• Iraqi Marshlands Project
• 
• IETC's Tools
• 
• ESTIS facilitates creation and management of websites on the Internet, sharing and searching of information across multiple ESTIS websites, publishing of information by non-web designers and decentralized management of content.
• 
• WiseWater is a spreadsheet application for projecting reductions in Water Consumption Patterns after application of Environmentally Sound Technologies. It is included as part of the publication "Every Drop Counts: Environmentally Sound Technologies for Urban and Domestic Water Use Efficiency".
• 
• PAMOLARE is an environmental modelling tool to forecast the changes in water quality leading to the eutrophication of Lakes and Reservoirs.
  The versatility of PAMOLARE allows for its use in decision making process as well as for training purposes.
•