Evapotranspiration data assimilation with genetic algorithms and SWAP model for on-demand irrigation

Evapotranspiration (ET) is one of the indicators of water use efficiency. Periodic information of ET based on remote sensing is useful for an on-demand irrigation (ODI) management. The main objective of this paper was to develop an ET data assimilation scheme to optimize the parameters of an agro-hydrology model for ODI scheduling. The soil, water, atmosphere, and plant (SWAP) simulation model has been utilized for this purpose. We computed remote sensing-based ET for a wheat field in the Sirsa Irrigation Circle, Haryana, in India using 18 cloud-free moderate resolution imaging spectroradiometer images taken between December 2001 and April 2002. The surface energy balance algorithm for land (SEBAL) was used for this purpose. Because ET estimates from SEBAL provide information on the surface soil moisture state, they were treated as observations to estimate unknown parameters of the SWAP model via a stochastic data assimilation (genetic algorithm) approach. The SWAP parameters were optimized by minimizing the residuals between SEBAL and SWAP model-based ET values. The optimized parameters were used as input to SWAP to estimate soil water balance for ODI scheduling. The results showed that the selected parameters (i.e. sowing, harvesting, and irrigation scheduling dates) were successfully estimated with the data assimilation methodology. The SWAP model produced reasonable states of water balance by assimilating ET observations. The root mean square of error was 0.755 and 2.132 cm³/cm³ for 0–15 and 15–30 cm soil depths the same layers, respectively. With optimized parameters for ODI, SWAP predicted higher yield and water use efficiency than traditional farmer’s irrigation criteria. The data assimilation methodology produced can be considered as an operational tool at the field scale to schedule irrigation or predict irrigation requirements from remote sensing-based ET.     

A simple visual aid for sugarcane irrigation scheduling

Irrigation scheduling based on soil water balance is a simple procedure that can be operated either manually or using computer programs. Adoption of the procedure is still low due to lack of soil water parameters and availability of climatic information. Furthermore, potential users are deterred by both the time and paper work required to carry out the calculations. In this study a visual device in the form of a plastic container was designed and tested to schedule sugarcane (Saccharum spp.) irrigation. The calibrated plastic bucket was field tested and proved to be an effective way to program sugarcane irrigation. It works simultaneously as a pluviometer and as an evaporimeter and, once it is marked, there is no need for human intervention beyond checking the position of the water level in relation to the irrigation control marks. It could be used with other crops and is useful for regions where meteorological data is scarce or difficult to obtain.     

Proposal for equitable water allocation for rotational irrigation in Pakistan

A rotational water supply system is designed to deliver a constant flow of water among irrigators along a tertiary canal. Under the existing rotational system in Pakistan transmission losses along the canal are not considered. A constant time per unit irrigated area is allocated to all the farmers regardless of their location along the canal. This results in decreasing volumes of water delivered to downstream farmers. A variable time model is developed which allocates more time to the downstream farmers to deliver a constant volume of water per unit area to all the farmers in the command area of a tertiary unit.     

Yield measures of irrigation performance in Pakistan

Customary evaluations of irrigation performance based upon crop yield per unit area do not reflect an adequate assessment of performance in water scarce environments such as Pakistan. Yield per unit water is a complementary and sometimes more appropriate measure. This note illustrates the different evaluations based upon yields of irrigated wheat and rice in Pakistan.     

基于SWAP模型的春玉米咸水非充分灌溉模拟

为了探究石羊河流域适宜春玉米生长的咸水非充分灌溉模式,应用SWAP模型模拟不同灌溉模式下的土壤水盐平衡、春玉米相对产量和相对水分利用效率,并预测了较长时期土壤水盐动态变化规律.研究结果表明:灌溉水矿化度为0.71 g/L和3.00 g/L的春玉米最优灌溉模式为生育期内灌4次水,灌溉定额均为408 mm,2种灌溉模式均能达到节约灌溉用水、提高作物产量和水分利用效率以及减少土体盐分累积量的目的.较长时期土壤水盐动态变化规律模拟结果表明:在冬灌条件下,春玉米最优灌溉模式下的土壤水分和盐分能够在模拟期内保持相对平稳的状态;在不同年份,相同土层土壤含水率随着土层深度的增加而增大,0.71 g/L的淡水灌溉土壤盐分主要累积在40~80 cm土层,3.00 g/L的微咸水灌溉土壤盐分主要累积在10~40 cm土层;5 a的模拟结果表明0.71 g/L和3.00 g/L的水持续灌溉5 a,不会引起土壤次生盐渍化.     

Water losses in smallholder irrigation schemes

Control of water losses is one important aspect of good water management. However, not all losses matter, and some are unavoidable. Many factors interact to determine the level of water losses, typically leading to very site-specific analysis. This paper discusses water losses in irrigated smallholder production systems, and raises the key issues concerning their control or minimisation. It explores the importance of different categories of loss, the significance of timing of losses, and the causes of losses. A rational analytical framework is proposed, in analogy with financial accounting and auditing, for identifying, quantifying, and evaluating the significance of losses, and arriving at a programme of remedial action where appropriate. The final sections of the paper draw some general conclusions about losses in different types of scheme, and make recommendations for relevant research.     

Factors affecting success and failure of trickle irrigation systems in Balochistan,Pakistan

An evaluation of performance of trickle irrigation systems installed in Balochistan, Pakistan during 1982–2002 was conducted by field surveys, physical verifications and interviews with farmers. Thirty systems were fully or partially operational and 76 had been abandoned. Successful systems required clean and reliable water supply, availability of spare components and accessories for replacements, skilled manpower, and a high level of interest and participation by the owner. The dominant species irrigated with these trickle systems were apples, grapes, and mixed orchards. Installations of trickle systems on old mature orchards were not generally successful due to lack of adaptation of the new system to limited and scheduled irrigation supplies. Many of the irrigators were not instructed on how to adjust the trickle system to meet changing needs of the plants. Consequently, growth of some of the trees was stunted and a few of them died. Lack of technical skill to repair and maintain the system and non-availability of replacement parts were general causes of failure of installed trickle irrigation systems. Clogging of the emitters was the primary specific cause of failure. Emitters with a larger opening that will not be clogged by most of the contaminants contained in the water available to these farmers and turbulent action screening systems to take out the other contaminants are proposed as solutions to this problem. Commercial shops, which sell the components, carry replacement parts and provide after-sales service are needed to keep trickle systems functioning in these isolated areas.     

Didactic tools for supporting participatory water management in collective irrigation schemes

In many irrigation schemes, Water Users Associations (WUAs) acquired the responsibility for water management after withdrawal of the state. Based on the success of some indigenous irrigation schemes, it was assumed that farmers could easily become managers. As decision-making was the exclusive terrain of the governmental agencies that ran the schemes, farmers never gained the necessary experience with water management. Therefore, training of farmers and WUAs on the principles and processes of water management is essential. This paper demonstrates a practical example of training material on water management that incorporates research results on the process of Irrigation Management Transfer and resulting farmers’ water management for the case of the Office du Niger irrigation scheme. The results of the paper explain how input from research was used in the four steps of the construction of training material. These steps are (1) setting the training agenda, (2) selecting and adapting information to be featured, (3) targeting the audience and (4) designing the actual training material. A first validation of the approach and examples of the actual training material were obtained in a workshop uniting all stakeholders.     

Effects of irrigation water salinity and electrostatic water treatment for sugarcane production

Excess salinity in irrigation water reduces sugarcane yield and juice quality. This study was conducted to compare the effect of irrigation with water of 1.3 dS m⁻¹ vs. 3.4 dS m⁻¹ on sugarcane yield and quality, and to evaluate whether an electrostatic conditioning treatment of the water influenced the salt effects. The study was conducted in a commercial field divided into large plots ranging from 1.0 to 1.2 ha in size. Cane and sugar yields were reduced approximately 17% by the 3.4 dS m⁻¹ water compared to the 1.3 dS m⁻¹ water, but juice quality parameters were not affected. Conditioning of the irrigation water using a device called an ‘electrostatic precipitator’ which claimed to affect various water properties had no effect on cane yield, juice quality or soil salinity levels. The detrimental effect of the high salt irrigation water was somewhat less than might be expected, probably due to good late summer rainfall which may have flushed the root zone from the excessive salts.     

Water productivity in smallholder irrigation schemes in South Africa

This study investigates the productivity and value of water in two smallholder irrigation schemes (Zanyokwe and Thabina) in South Africa. We apply the residual valuation method (RVM), willingness to pay (WTP) and cost-based approaches (CBA) (i.e. accounting costs of operation and maintenance) to evaluate water productivity and values per crop, per farm, and by scheme. In both schemes, water value estimated by the RVM for vegetables (cabbage, tomatoes and butternuts) is greater than water value for dry maize. At the farm and scheme levels, a comparison was made between gross margin per m³ of water, WTP per m³ and accounting cost per m³ to estimate the relative value of water productivity. The active farmers in the Zanyokwe scheme have lower WTP per m³ (R0.03) than the gross margin of output (R0.69). Also, the accounting cost (R0.084) per m³ of water is less than the gross margin. In the Thabina scheme, active farmers are willing to pay (R0.19) per m³ of water. Hence farmers in Thabina are ready to pay as much as three times the proposed costs of O&M (R0.062) per m³ of water used. Both the accounting cost and willingness to pay are less than the gross margin per m³ of water in the Zanyokwe scheme. Our findings indicate that extension and training may be required to improve the productive use of water for those farmers whose returns are insufficient to cover the cost of supply.     

Evaluation of fertigation scheduling for sugarcane using a vadose zone flow and transport model

Micro-irrigation has become an optimal means for providing water and nutrients to crops. There is an ample space for improving fertilizer use efficiency with micro-irrigation, if the movement and reactions of fertilizers in the soil are well understood. However, the rhizosphere dynamics of nutrients is very complex, depending on many factors such as soil temperature, pH, water content, and soil and plant characteristics. Many factors cannot be easily accurately quantified. However, using state-of-the-art modelling techniques, useful and reliable information can be derived.An attempt was made to evaluate the reactive transport of urea in the root zone of a sugarcane crop under drip irrigation, and to quantify the fluxes of urea, ammonium, and nitrate into the crop roots, volatilization fluxes, and deep drainage using a numerical model. This quantification helped in designing an optimal fertigation schedule. Various parameters used in the model were taken from either the literature or the field study. A typical scenario, based on the recommended total quantity of urea for sugar cane crop under drip irrigation in India, was tested using HYDRUS-2D. The total amount of urea was divided into fortnightly doses, depending on the stage of crop growth. For this scenario, the modelled crop uptake was found to be 30% higher than the crop demand. Consequently, an optimal fertigation schedule was developed that reduced the use of urea by 30% while at the same time providing enough N for its assimilation at all stages of crop growth. This type of modelling study should be used before planning field experiments for designing optimal fertigation schedules.     

Performance assessment of irrigation water management of heterogeneous irrigation schemes: 1. A framework for evaluation

Studies of the performance assessment of irrigation schemes have gained momentum since the late 1980s due to the common perspective that the resources (land and water) in irrigation schemes are not being managed appropriately. In this paper irrigation water management is considered as one of the activities of the irrigation scheme. Three phases of irrigation water management namely planning, operation and evaluation are identified. A framework for the performance assessment of irrigation water management in heterogeneous irrigation schemes is proposed in this paper, based on earlier studies made in this direction. The paper presents two types of allocative measures (productivity and equity) and five types of scheduling measures (adequacy, reliability, flexibility, sustainability and efficiency), together with the methodologies for estimating these for the scheme as a whole during different phases of irrigation water management.     

The fate of nitrogen applied to sugarcane by trickle irrigation

Fertigation can be a more efficient means of applying crop nutrients, particularly nitrogen (N), so that nutrient application rates can be reduced in fertigated crops. However, there is little information on the extent of the possible reduction in N application rate for fertigated sugarcane, one of the major row crops grown under trickle irrigation, nor the fate of N in fertigated sugarcane systems if N application rates are not reduced. An experiment was established to determine the response of cane and sugar production to different N rates (0–240 kg ha^–1 year^–1) spanning that recommended for conventional irrigation systems (160 kg ha^–1 year^–1). As well as yield, N removed in the crop and changes in soil mineral N were determined annually for four crops (a plant and three ratoon crops). ¹⁵N values were also measured in selected treatments at selected times to assess possible N inputs to the experiment via biological N fixation (BFN). Yields of cane and sugar responded to application of N fertiliser in the three ratoon crops, but they were not significantly increased by applying more than 80 kg ha^–1 of N. There were no N responses in the plant crop, as there was >200 kg ha^–1 of soil mineral N (SMN) to 2 m depth at the site prior to planting, and much of this SMN was depleted in the treatment receiving no N. There was no evidence of N input from BFN in the experiment. During the 4-year study period, net removal of N from the treatment with no applied N totalled 207 kg ha^–1. When 80 or 120 kg ha^–1 year^–1 of N was applied to ratoon crops, outputs of N from the harvested crop approximately balanced inputs from fertiliser and depletion of SMN during the experiment. Inputs clearly exceeded output at higher N application rates. Assuming that the net removal of N from the treatment with no applied N was the same as the net mineralisation of N from soil organic matter in all treatments in the experiment, 204–639 kg ha^–1 of N was unaccounted for in the treatments with applied N over the duration of the experiment. While some of this N (e.g. 45 kg ha^–1) may have resulted in small (and undetectable) increases in total soil N, much of it would have been lost to the environment. We suggest that the high soil water contents maintained with daily application of irrigation water through the trickle system promotes mineralisation of soil organic matter and hence losses of N to the environment. Thus, particular care is required to avoid over-application of N in fertigated sugarcane.Communicated by K. Bristow     

Evaluation of on-farm irrigation applications using the simulation model EPIC

An understanding of water needs in agriculture is a critical input in resolving the water resource issues that confront many southeastern and other US states. The objective of this study was to evaluate on-farm irrigation applications for three major crops grown in Georgia, USA using the Environmental Policy Integrated Climate (EPIC) model. For cotton, 16, 58, and 75 farmers fields in 2000, 2001, and 2002, respectively, were selected from among the Agricultural Water Pumping (AWP) program sites across the state of Georgia. For maize, 9, 20, and 28 fields were selected in 2000, 2001, and 2002, respectively, and for peanut, 18, 51, and 54 fields were selected in 2000, 2001, and 2002, respectively. The majority of these fields were located in the southwest region of Georgia, where traditional row-crop agriculture is most dominant. We compared the simulated irrigation requirements with the amount of water that the farmers actually applied during the 2000, 2001, and 2002 growing seasons. For cotton and peanut, the means of farmer-applied irrigation amounts and simulated irrigation requirements agreed very well, with similar values for root mean squared deviation (RMSD) of the two crops. For maize, good agreement between simulated and farmer-applied irrigation amounts were found only in 2001. Farmers applied more water to their maize crop when compared to simulated irrigation requirements, especially when rainfall was very low and potential evapotranspiration was high during the 2000 and 2002 growing seasons. The component of the mean squared deviation (MSD = RMSD²) related to the pattern of variability in seasonal irrigation applications contributed most to MSD. Accurate estimates of the mean and the magnitude of variability in seasonal irrigation applications could be very useful for the estimation of overall water use by agriculture in Georgia and other southeastern states. This study showed that the EPIC model would be an adequate tool for this purpose; potential users could include policy makers, planners and regulators, including the Georgia Department of Natural Resources (DNR).     

Model for performance based land area and water allocation within irrigation schemes

This paper focuses on irrigation schemes under rotational water supply in arid and semiarid regions. It presents a methodology for developing plans for optimum allocation of land area and water, considering performance measures such as productivity, equity and adequacy. These irrigation schemes are characterized by limited water supply and heterogeneity in soils, crops, climate and water distribution network, etc. The methodology proposed in this paper, therefore, uses a previously developed simulation–optimization model (Area and Water Allocation Model, AWAM) that considers the heterogeneity of the irrigation scheme in the allocation process, and modifies this to take account of equity and adequacy of supply to irrigated areas. The AWAM model has four phases to be executed separately for each set of irrigation interval over the irrigation season: 1. generation of irrigation strategies for each crop–soil–region combination (CSR unit), 2. preparation of irrigation programmes for each irrigation strategy, 3. selection of specified number of irrigation programmes for each CSR unit and 4. optimum allocation of land area and water to different parts of the irrigation scheme (allocation units) for maximizing productivity. In the modified AWAM model, the adequacy is included at Phase-2 (by including only the irrigation programmes for full irrigation of each CSR unit) and equity is included at Phase-4 (by including the constraints for equity). The paper briefly discusses the applicability of the modified AWAM model for a case study of Nazare medium irrigation scheme in Southern India. The results of the case study indicated that the performance measures of productivity, equity and adequacy conflict with each other.     

Evaluation of infiltration measurements for border irrigation

A number of methods are discussed for obtaining a reasonable estimate of the infiltration function for irrigation borders. Data from ring infiltrometers are fit to power functions for infiltration rate and cumulative infiltration rate versus time and to a branch function where the infiltration rate is not allowed to go below some value (called the final infiltration rate). A volume balance within the border is used to adjust the data to give a better indication of the “average” infiltration conditions over the border. The results of Bouwer's method, which uses a series of borders as infiltrometers, were compared to the results of ring data for actual field data. Bower's method was also analyzed by developing advance and recession curves with the zero-inertia border-irrigation model with a known infiltration rate. The zero-inertia model was also used to examine the effect of different infiltration functions for specific examples (resulting from different irrigations or different estimation methods) on the application of water by surface irrigation.     

Towards using a thermal infrared index combined with water balance modelling to monitor sugarcane irrigation in a tropical environment

In humid regions, the timing and quantity of a complementary irrigation regime is challenging because of the irregularity of rainfalls events. In this study, we tested the use of a thermal infrared derived empirical crop water stress index (CWSI_e) as an in situ measurement of the water status of sugarcane, to better monitor the irrigation scheduling. To do this, we set up a 2-year experiment in Reunion Island, on a trial with plots under different water conditions (rainfed and irrigated). Crop surface temperature was measured daily with infrared radiometers (Apogee Instruments) installed above the canopy, and soil moisture and drainage measurements were used to derive the ratio between actual and maximum evapotranspiration (AET/MET) values that were then averaged on “hydrically homogeneous” time periods (between 7 and 25 days). Only the thermal data acquired on clear days and 1 h after noon in 2007 were used to define the empirical lower and upper baselines required for the calculation of empirical CWSI. The data set acquired in 2008 was used to test the robustness of the method as we used the upper and lower baselines defined in 2007 to calculate CWSI_e. The linear regression between AET/MET and (1 − CWSI_e) averaged on the same periods (values ranging between 0.4 and 1) showed a significant correlation for both experimental years (global R² = 0.75 and RMSE = 0.12). This result indicates the effectiveness of the CWSI_e to measure the water status of the sugarcane crop, even in humid conditions with a vapor pressure deficit (VPD) between 0.5 and 2.1. We conclude the study by discussing the complementarity of this remote water stress index (CWSI_e) with OSIRI water balance modelling tool currently used in Reunion Island for monitoring sugarcane crop irrigation.     

An evaluation of the performance of six smallholder irrigation schemes in Zimbabwe

Informal and formal survey procedures were used to assess the performance of six smallholder irrigation schemes. The predominant crops grown at the schemes were maize (Zea mays) in summer and beans (Phaseolus vulgaris) in winter. Average crop yields range from 2.7 to 7.4 t/ha for maize and 0.4 to 2.4 t/ha for beans at the lowest and highest yielding schemes respectively. There was also a large gap between the lower and higher producing farmers at each scheme. Only one out of the six schemes realized a profit margin of more than ZWD 223 per month per plotholder (the minimum wage rate for agriculture workers) after operating and maintenance costs (currently met by government) were deducted. Major constraints on the schemes included limited cash for input purchases, unreliable water supplies for winter cropping, limited market outlets and poor roads. Ways of alleviating some of these constraints to improve overall scheme performance are proposed.     

Performance assessment of irrigation water management of heterogeneous irrigation schemes: 2. A case study

The non-uniformity of soils, weather, fields, cropping pattern and canal systems in most surface irrigation schemes makes irrigation water management complex, but optimum performance is important particularly in irrigation schemes with limited water supply. This paper focuses on the performance of irrigation water management during the area and water allocation with a case study of an irrigation scheme in the semi-arid region of India. Often the irrigation managers or authorities of these heterogeneous irrigation schemes also need to deal with different allocation rules. The allocation plans and the corresponding water delivery schedules during the allocation process were estimated with the help of a simulation–optimisation model for different allocation rules based on cropping distributions (free and fixed), water distributions (free and fixed-area proportionate), irrigation depth (full, fixed depth and variable depth irrigation) and irrigation interval (from 14 to 35 days). The performance measures of productivity (in terms of net benefits and area irrigated), equity (in water distribution), adequacy and excess were assessed for these different allocation plans and schedules. These were further compared with the performance measures of the existing rule (fixed depth irrigation at a fixed interval). The analysis revealed that these performance measures are in some cases complimentary and in other cases conflicting with each other. Therefore, it would be appropriate for the irrigation managers to understand fully the nature of the variation in performance measures for different allocation rules prior to deciding the allocation plans for the irrigation scheme.     

WaDI (water delivery for irrigation): A simulation tool to address strategic interaction of water demand and supply in irrigation schemes

The WaDI (water delivery for irrigation) model deals with the relations between the collective water supply and the demand within irrigated schemes. It is based on a separate modelling of the water supply and demand, including (i) a simplified representation of the hydraulic structure, characteristics and organizational parameters of the scheme, (ii) an assessment of the water demand of farms based on their total irrigated area, cropping pattern and irrigation practices, (iii) a farm typology, (iv) a confrontation between water demand and supply at each node of the scheme from pumping plants to tertiary canals, and (v) calculation of response factors between supply and demand during the peak demand period. “What-if” scenarios are simulated in order to enhance the stakeholders’ capacity to plan strategic decisions such as water delivery allocation rules or infrastructure investments. WaDI was implemented on two Brazilian schemes. It allowed broadening the stakeholders’ limited representation of collective water supply and demand into a more comprehensive understanding of these relations. The approach however showed some limits, along with the difficulty of assessing the real impact on the stakeholders’ capacity for strategic planning.