Evapotranspiration and water use of full and deficit irrigated cotton in the Mediterranean environment in northern Syria

https://doi.org/10.1016/j.agwat.2011.02.009Get rights and content

Abstract

Cotton (Gossypium hirsutum L.) is the most important industrial and summer cash crop in Syria and many other countries in the arid areas but there are concerns about future production levels, given the high water requirements and the decline in water availability. Most farmers in Syria aim to maximize yield per unit of land regardless of the quantity of water applied. Water losses can be reduced and water productivity (yield per unit of water consumed) improved by applying deficit irrigation, but this requires a better understanding of crop response to various levels of water stress. This paper presents results from a 3-year study (2004–2006) conducted in northern Syria to quantify cotton yield response to different levels of water and fertilizer. The experiment included four irrigation levels and three levels of nitrogen (N) fertilizer under drip irrigation. The overall mean cotton (lint plus seed, or lintseed) yield was 2502 kg ha−1, ranging from 1520 kg ha−1 under 40% irrigation to 3460 kg ha−1 under 100% irrigation. Mean water productivity (WPET) was 0.36 kg lintseed per m3 of crop actual evapotranspiration (ETc), ranging from 0.32 kg m−3 under 40% irrigation to 0.39 kg m−3 under the 100% treatment. Results suggest that deficit irrigation does not improve biological water productivity of drip-irrigated cotton. Water and fertilizer levels (especially the former) have significant effects on yield, crop growth and WPET. Water, but not N level, has a highly significant effect on crop ETc. The study provides production functions relating cotton yield to ETc as well as soil water content at planting. These functions are useful for irrigation optimization and for forecasting the impact of water rationing and drought on regional water budgets and agricultural economies. The WPET values obtained in this study compare well with those reported from the southwestern USA, Argentina and other developed cotton producing regions. Most importantly, these WPET values are double the current values in Syria, suggesting that improved irrigation water and system management can improve WPET, and thus enhance conservation and sustainability in this water-scarce region.

Highlights

Deficit irrigation has moderate impact on cotton water productivity. ► Rationalizing water application can contribute to natural resources conservation. ► Sufficient fertilization can contribute to natural resources conservation in water scarce areas.

Introduction

Cotton (Gossypium hirsutum L.) is grown in many countries, under both rainfed and irrigated conditions. In the semi-arid Mediterranean region, in places like southern Spain, northern Syria and western Turkey, the crop is entirely irrigated (Janat and Somi, 2001) with very limited rainfall during the summer growing season. Cotton requires large quantities of water, while water supplies are declining. Clearly, there is an urgent need for technical options as well as policy measures to encourage environmentally sustainable, yet economically viable practices. This study focuses on Syria, where cotton is vital to the national economy, while simultaneously, shallow aquifers are being depleted at an alarming rate. Water is increasingly pumped from deeper groundwater and the two most important production inputs – irrigation and fertilizers – are becoming the two largest production costs (Janat and Somi, 2001).

Cotton production in Syria dates back to ancient times. Cotton is economically more competitive than any other summer crop. It is a major source of income for one-fifth of the country's economically active population. Most fields range from 2 to 25 ha. Cotton is grown in a two year rotation with wheat, sugar beet, potato, legumes, and vegetables. Typically, wheat is harvested in June; land is plowed twice and left fallow. A third plowing is done in February or March of the following year, and cotton planted in April and May. Planting is mostly by hand, and quality seeds are available from the government. The harvest starts in September and lasts through December; the entire crop is handpicked in two or more pickings. Water is conveyed to the fields mainly through earthen ditches. The most common practice (seen on 75% of all cotton fields) is flooding of small basins, followed by furrow irrigation (24% of the fields). Irrigation applications are double (or even higher) the crop requirements, with a national average of about 1500 mm per season. Cotton accounts for an estimated 25% of Syria's total agricultural water use.

The national average for seed cotton yield is around 4 t ha−1. Productivity per unit area is at acceptable levels, but yield per unit of applied water is extremely low because of inefficient irrigation systems and improper water management. On a global level, acceptable yield of irrigated cotton is 4–5 t ha−1 seed cotton with water productivity (WPET) values of 0.4–0.6 kg per m3 of depleted water. This range is also inferred from the recent reviews of Grismer (2002) and Zwart and Bastiaanssen (2004), for data from multiple countries. Average WPET was reported as 0.65 and 0.23 kg m−3 for seed cotton and lint cotton, respectively, with a large variability ranging from 0.41 to 0.95 kg m−3 for seed cotton and 0.14–0.33 kg m−3 for lint cotton. The large variability is due to many factors, including differences in climate, soil, and irrigation and nutrient management, but suggests opportunities for maintaining or increasing production with less water. In Syria, WPET in the traditional surface irrigated cotton is only 0.2–0.25 kg m−3 of seed cotton and 0.07–0.09 kg m−3 for lint cotton. These values are only half to one-third of those achieved in most major cotton producers such as Argentina, Turkey, and USA (Hunsaker et al., 1998, Ayars et al., 1999, Howell et al., 2004, Dagdelen et al., 2006). They are also lower than the minimum values reported in global reviews (Grismer, 2002; Zwart and Bastiaanssen 2004).

More than 60% of irrigated lands in Syria use groundwater (Salman, 2004) and are not part of government irrigation schemes. These irrigators enjoy on-demand irrigation water, and thus potentially have the flexibility and control to implement an effective on-farm water management through scheduling, reallocation and other means. Currently, most farmers aim to maximize yield (and presumably profit) by maximizing irrigation. This is both unwise and unsustainable in basins where water is being withdrawn faster than it is being replenished. A better alternative is a targeted demand management strategy that may include water rationing or deficit irrigation (Farahani et al., 2006). Deficit irrigation, either voluntarily or regulated, is an option that may increase WPET (Kijne et al., 2002), and would most certainly improve resource sustainability.

Cotton is an indeterminate perennial shrub that is suitable for conditions of limited water and tolerant to salinity. Past research, dating back to at least the 1930s (DeTar, 2008), documents various aspects: physiological and morphological responses to water, deficit irrigation and its economics (English et al., 1990), and water use and yield relationship (Wanjura et al., 2002, Howell et al., 2004, Dagdelen et al., 2006). Drip irrigated cotton data from Wanjura et al. (2002) show reduced yields for deficit as well as for over-irrigation. Falkenberg et al. (2007) reported that irrigation at 75% of ETc did not reduce cotton yield. Data from Howell et al. (2004) and DeTar (2008) do not show any gains in WPET due to deficit irrigation. For a range of irrigation regimes starting at about 50% of the optimum application of 654 mm, DeTar (2008) observed reduced yields due to deficit irrigation. At the optimum application, WPET averaged 0.219 kg m−3 for lint cotton, which was reduced in deficit irrigated plots, as well as in over-irrigated plots where application exceeded requirements by 30%. However, past research clearly shows cotton yield reductions due to excess water (Wanjura et al., 2002 and Karam et al., 2006). The literature reports mixed results on the impact of water rationing on WPET of cotton, but a reduction in WPET due to over-watering is most certain. An increase in WPET due to deficit irrigation is neither obvious nor universally observed as it is a complex interaction of many factors including timing and duration of the stress in addition to variations in application methods and efficiencies.

Literature from Syria includes comparison of irrigation methods and fertilizers on yield, but no study of deficit irrigation is available. Local production functions are needed, especially since results from literature are mixed and difficult to transfer with certainty. To determine yield response to water, precise implementation of irrigation regimes and accurate measures of crop water use (ETc) and yield are needed. This was the objective of this study that was implemented in drip-irrigated cotton in northern Syria under four levels of water rationing regimes and three levels of nitrogen fertilizer during the period 2004–2006. Results quantify yield response to water and are useful for potential development of water demand management strategies and economic analysis.

Section snippets

Site description and field practices

The field study examined the effects of varying soil water and fertility regimes on production and water productivity of drip-irrigated cotton (Gossypium hirsutum L.) in the typical Syrian practice of two-year cotton-wheat rotation. This study was conducted at Tel Hadya research station (36°01′N, 36°56′E, and 284 m above mean sea level), the headquarters of the International Center for Agricultural Research in the Dry Areas (ICARDA), located 35 km south of Aleppo in northern Syria. The study was

Rainfall and temperature

The climate during the three growing seasons was very similar (Table 3). There was practically no rainfall during the three months of June, July and August, and only a few millimeters of rain in May and September. The monthly Class-A pan evaporation and total for the three seasons show little variations with an average value of 1841 mm over the three seasons. These variations are typical for the Mediterranean climate that is characterized by a long hot and dry summer, during which full

Discussion

Water productivity is computed as yield per ETc (WPET) and yield per applied irrigation water (WPiw). The former is more a biological indicator while the latter is influenced by the performance of the irrigation system and the degree of water losses beyond transpiration. Mean WPET decreased with increasing water rationing (from more than 0.39 kg m−3 at 100% irrigation to about 0.32 kg m−3 at 40% irrigation) while WPiw increased with increasing water rationing (from 0.43 at 100% irrigation to 0.48 kg m

Conclusions

Both yield and water productivity (WPET) of cotton increase with an increase in irrigation level. This was most pronounced at optimum N fertility levels. From a biological view point, deficit irrigation may not necessarily improve water productivity. However, economically, the conclusion may be different depending primarily on the cost of irrigation and/or water and the sale value of the produce. The WPET values obtained in this study are double the current cotton WP values in Syria, suggesting

Acknowledgements

The authors wish to thank Pierre Hayek, Jihad Abdullah, Ali Haj Dibo, Issam Halimeh, and other ICARDA field staff for managing the field trials and for carrying out the tedious and labor intensive soil water content and plant measurements. Appreciation is also extended to Dr. Murari Singh for his generous help with statistical design and data analysis.

Cited by (0)

View full text