Sensitive variables controlling salinity and water table in a bio-drainage system
- First Online:
- Cite this article as:
- Akram, S., Kashkouli, H.A. & Pazira, E. Irrig Drainage Syst (2008) 22: 271. doi:10.1007/s10795-008-9056-4
- 104 Views
Bio-drainage can be considered as an important part of sustainable irrigation water management. Bio-drainage has potential for managing shallow water conditions in arid and semiarid areas especially when traditional subsurface drains are not available. Bio-drainage theory does not go back too far. The relationship between soil characteristics, water management regimes, and climatic conditions is not yet well defined. This study attempted to use a mathematical model (SAHYSMOD) to evaluate factors affecting design and operation of a bio-drainage system and study its sensitivity to different variables. The study showed that the major constraint of bio-drainage is salt accumulation in tree plantation strips in arid and semiarid regions. Maximum soil water salinity which can be controlled by bio-drainage is around 3 dS m−1 in rather medium run and sustainability may only be achieved where a salt removal mechanism is considered. The study also showed that the effectiveness of the system is higher where the neighboring strips are narrower. It also showed that bio-drainage is very sensitive to the amount of applied water. While the barrier depth does not have an important effect on water table draw down, it does have a great influence on lowering the salinization rate of tree plantation strips. The application of bio-drainage could be economically controversial since in humid areas water is sufficient for agricultural crops, allocating parts of the expensive land to mostly non-fruit trees may not be feasible, while in arid and semiarid regions there is usually enough cheap land to grow trees.
Drainage can be considered as an important part of sustainable and Integrated Water Resources Management. Drainage water in arid and semi arid regions is usually saline, and sometimes incorporated with nutrients, toxic materials left from residues of pesticides and herbicides, as well as trace elements. These substances are not suitable for the environment; their disposal to surface water such as rivers, lakes, and wetlands might be quite harmful. Nontraditional drainage methods could be utilized to maintain shallow water table and soil salinity at harmless levels, and prevent pollutant disposal to the surface aquatic bodies.
Among the main environmentally friendly drainage systems are bio-drainage, dry drainage, controlled drainage and agroforestry. Bio-drainage is a natural system, in which tree plantation strip absorbs deep percolation losses of irrigation water applied to the neighboring crop strip and dispose excess water through evapotranspiration. In other words the concept of bio-drainage is based on evapotranspiration from tree plantation strips located adjacent to the irrigated crop strips (Chhabra and Thakur 1998, Liaghat 2004). Hydraulically speaking, the crop strips and the tree plantation ones are “sources” and “sinks”, respectively. The method could be economically feasible in some parts of the world since it only needs a fixed capital investment at the beginning, while it produces fiber, wood, and animal fodder in successive years.
It is believed that bio-drainage technology is capable of maintaining water table at a desired level and to some extent relieves waterlogging and canal seepage problems, if designed properly (Smedema 1997). It is doubtful, however, if it can maintain soil salinity to an extent that crops could be grown economically (Heuperman et al. 2002). In saline environments, hybrid systems that combine bio-drainage and conventional engineering-based technology will be needed to achieve sustainability. Under this scenario the bio-drainage component improves overall system efficiency and minimizes the high-capital inputs associated with engineering-based drainage methods (Heuperman 2003).
According to Kapoor and Denecke (2001) bio-drainage could be used in various regions ranging from humid to semi arid areas, except when the ground water EC is greater than 12 dS m−1. Compared to traditional drainage systems i.e. horizontal and vertical, bio-drainage is the most cost effective and environmentally friendly option. Using bio-drainage, the requirements of conventional drainage systems such as field drains, conveyance systems, effluent disposal facilities, and pumping stations could be eliminated. In most cases bio-drainage could have the first priority compared with horizontal relief drains and/or vertical tube wells. The main constraint of bio-drainage is the need for salt removal and extra land for tree plantation. In semiarid areas this is not a real limitation since available irrigation water usually does not suffice the land available.
In bio-drainage studies the designer must be concerned about a broad range of variables affecting water balance, salt balance, area to be allocated to plantation, plant species, consumptive use of plants, groundwater quality, soil characteristics, and more importantly the ratio of the plantation width to the width of the crop strip (Akram 2006). Plant species should be selected from among high evapotranspirative, salt resistant deep rooted ones, such as Tamarix troupii, Acacia tortilis, Acacia nilotica, and different eucalyptus species, particularly Eucalyptus camaldulensis.
Heuperman et al. (2002) mentioned two major conclusions: Bio-drainage can effectively contribute to strongly reducing the problems as experienced from waterlogging in irrigated agriculture and non-irrigated agriculture and the problems associated with a rise in salinity in the root zone can be effectively delayed using bio-drainage systems in semiarid and arid areas.
Materials and methods
Needs to use a mathematical model
Numerous variables affect performance of bio-drainage (Akram 2006, Heuperman et al. 2002, Singh et al. 2002). Field experiments need quite different types of soils (both from physical and chemical points of view), climates, water quality, crop and tree species having different depths of irrigation and evapotranspiration, respectively, and above all several years time to monitor the results for different widths of crop and plantation strips. In other words it is almost impossible to do the research in the field to evaluate the effects of so many parameters on the performance of bio-drainage. So, it seems that a mathematical model is the most suitable way to simulate water table, and salt accumulation in the soil simultaneously, both for crop and plantation strips in different soil and climate conditions. No doubt the most promising results should be tested in the field.
Figure 1 shows that in bio-drainage there is a vertical flow i.e. irrigation water applied to crop strips and evapotranspiration from adjacent tree planted strips, and a horizontal flow between the neighboring strips. So, at least a 2D model is needed. SAHYSMOD simulation model was found to be an appropriate one. It combines the agro-hydro-salinity model, SALTMOD (Oosterbaan 2002), and the nodal ground water model (Standard Groundwater Model Package (SGMP), developed by Boonstra and de Ridder 1990) which was made by K.V.G.K. Rao. The calculation programmes were elaborated in FORTRAN and the user shell in Turbo Pascal (SAHYSMOD working group of ILRI 2003).
SAHYSMOD is a computer program for the prediction of the salinity of soil moisture, ground and drainage water, the depth of the water table, and the drain discharge in irrigated agricultural lands, using different geohydrologic conditions, varying water management options, including the use of ground water for irrigation, and several cropping rotation schedules, whereby the spatial variations are accounted for through a network of polygons. SAHYSMOD uses Block Centered Finite Difference method to solve the well known Bousinesque equation.
If one wishes to determine the effect of variations of a certain parameter on the value of other parameters, the program must be run repeatedly according to a user-designed scenario. This procedure is used here in this research. It must keep in mind that although ecohydrological models offer predictions for the future, they may become inaccurate due to over- or under parameterization (Singh 2005).
SALTMOD, SGMP, and SAHYSMOD have all been evaluated in the past and have proven their credibility (Shirahatti et al. 2001).
An agronomic water balance model, which calculates for each polygon the downward and upward water fluxes in the soil profile (SALTMOD);
A groundwater model of the aquifer, which calculates the groundwater flows into and from each polygon and the groundwater levels per polygon (SGMP); and
A salt balance model which runs parallel to the water balance model and determines average EC in the soil profiles of both crop strips and tree planted strips which are carried by water from crop strips, and then evapotranspirated from tree plantation strips. This study is confined only to the transfer of Total Dissolved Solids (TDS) represented by EC, and not to any specific salt.
Seasonal approach based on daily calculations
– A surface reservoir(s);
– An upper soil reservoir in the root zone (r) which can be saturated, unsaturated, or partly saturated, depending on the water balance. All water movements in this zone are vertical, either upward or downward, depending on the water balance. SALTMOD takes care of the vertical flow;
– A transitional soil reservoir zone (x) which can also be saturated, unsaturated or partly saturated. All flows in this zone are horizontal, except the flow to traditional subsurface drains, which is radial. SGMP takes care of the horizontal flow between and amongst the neighboring polygons; and
– A deep reservoir or main aquifer.
Subsurface drainage water used for irrigation (cubic meters per square meter total polygonal area);
Outgoing surface runoff or surface drain water from irrigated land (cubic meters per square meter irrigated area);
Total actual evapotranspiration (cubic meters per square meter total area);
Total amount of subsurface drainage water (cubic meters per square meter total area);
Horizontally incoming ground water flow into a polygon through the aquifer (cubic meters per day); and
Horizontally outgoing ground water flow from a polygon through the aquifer (cubic meters per day);
Ground water pumped from wells in the aquifer (cubic meters per square meter total area);
Gross amount of field irrigation water (cubic meters per day per square meter total area);
Gross amount of outgoing field irrigation water (cubic meters per day per square meter total area);
Percolation from the irrigation canal system (cubic meters per square meter total area);
Percolation from the root zone (cubic meters per square meter total area);
Infiltration through the soil surface into the root zone (cubic meters per square meter non-irrigated area);
Rainfall/precipitation (cubic meters per square meter total area);
Vertical downward drainage into the aquifer (cubic meters per square meter total area); and
Velocity of vertical upward seepage from the aquifer (meters per day).
Agricultural water balances
Ground water flow
Annual input changes
The program can be run either with fixed input data for the number of years determined by the user or with variable input data. The first option was used to predict future developments based on long-term average input values, as it will be difficult to assess the future values of the input data year by year.
- Soil and water:
four values for hydraulic conductivity, i.e. K = 0.25, 0.5, 1.0, and 2.0 m day−1;
five values for depth to the barrier, i.e. d = 2, 4, 6, 8, and 10 m;
two values for initial water table depth in crop strips, i.e. 0.2, and 0.4 m;
four values for initial irrigation water salinity, i.e. C1 = 1, 2, 5, and 10 dS m−1; and
four values for initial average soil salinity, i.e. C0 = 1.5, 3, 7.5, and 15 dS m−1.
Recharge and discharge:The study is done without considering any particular crop in crop strips, using different drainage rates instead. The same is true for plantation strips in which different values for evapotranspiration were considered.
Three values for drainage rate, i.e. q = 1, 2, and 3 mm day−1;
One value for annual precipitation, i.e. 250 mm/year equally distributed; and
Two values for evapotranspiration rate from plantation strips i.e. 3,000, and 4,000 mm/season. Although 4,000 mm/season seems to be quite a high value, it was included to verify if bio-drainage can still work under this extreme condition without crucial impairment of the trees due to salt accumulation.
- Land layout:
Parallel strips, with high ratio of their lengths to their widths in order to minimize the effect of the boundaries;
19 values for strip widths as shown in Table 1.
Different combinations of plant and crop widths
Crop strip width (m)
Plant strip width (m)
The average soil salinity of crop strips evenly distributed in the soil profile does not exceed 8 dS m−1 after 20 years, in which still some annual crops such as alfalfa, barley and sugar beets could be grown;
The average soil salinity of plantation strips uniformly distributed in the soil profile does not exceed 32 dS m−1 after 20 years, which a few trees and bushes such as species of eucalyptus, acacia, atriplex, and tamarix could still tolerate;
The water table in crop strips is not shallower than 0.5 m, suitable for some annual shallow rooted crops;
The water table in the midpoint of plantation strips is not shallower than 1.2 m;
In case where several plantation widths (Lp) satisfy the above criteria for a given crop width (Lc), the one with minimum Lp/Lc ratio i.e. the more economical one would be accepted.
Results and discussion
The effect of hydraulic conductivity on water table
Figure 3a also shows that very low values of hydraulic conductivity may not be able to draw water table down to the desired level in crop strips. This result is quite the same as what one can expect from conventional drainage systems. However, the problem may not be important when Lp/Lc has higher values. In almost all conditions when the hydraulic conductivity exceeds 1 m day−1, the water table does not fall any more in crop strips. So, bio-drainage is less effective in heavier soils, hence, more attention should be paid in its design. Higher values of Lp/Lc, however, will solve the problem, but of course with the expense of allocating more land to plantation area.
The effect of hydraulic conductivity on soil salinity
Increasing hydraulic conductivity does not have a significant role in increasing average salinity either in crop strip or in plantation strip since the water movement occurs due to potential difference created by evapotranspiration from tree plantation strip. This is true at least for q < 3 mm day−1. The lower the Lp/Lc ratio, the higher the salt concentration in plantation strips. This is due to salt balance, in which the salt mass transferred to the plantation strip remains constant while the plantation strip width decreases, and consequently the salt concentration increases.
The effect of barrier depth on water table
The effect of barrier depth on salinity of the plantation strip
Increasing the depth to the barrier causes less salt concentration in plantation strips. This is due to salt balance, in which the salt mass transferred to the plantation strip remains constant. While the volume of the soil increases, the salt concentration must decrease. This means that bio-drainage is more feasible and can last longer in areas where the impervious layer is deeper.
The higher the Lp/Lc ratio, the lower the salt concentration in plantation strips. This means that with allocation of more land to tree plantation, the efficiency and the life of the bio-drainage system increases.
Sustainability of bio-drainage
Sustainable projects are those which save the physical, ecological, social, and economic environment in the long run. It is expected, however, that so called environmentally friendly drainage methods be really sustainable without any harm to the environment. Most authors believe that bio-drainage is able to maintain water table in a position not to be hazardous to the crop. However, some are doubtful about its sustainability from the point of view of soil salinity. In fact, according to the salt balance principle, salt accumulation in strips having upward flow, i.e. plantation strips, is inevitable unless some type of salt removal mechanism such as foliage harvesting, salt scraping from the soil surface, and/or leaching facility exists. Heuperman et al. (2002) and Heuperman (2003) suggest a combination of bio-drainage and a conventional one.
Bio-drainage and salinity in crop strips
Bio-drainage and salinity in plantation strips
Bio-drainage effectiveness and the widths of the strips
The results show that with a constant Lp/Lc ratio, the effectiveness of the system is higher when both Lp and Lc have smaller values. For example, when Lp/Lc = 1, the water draw down is higher and salt accumulation is lower when Lp = Lc = 25 m compared to when Lp = Lc = 75 m. The designer should therefore take the narrower strips into consideration to such an extent that it does not impair the work of agricultural machinery and does not make any other limitations.
High sensitivity of bio-drainage to salinity in arid and semi arid regions
Suitable regions for bio-drainage
Bio-drainage does have a high sensitivity to salinity in regions with arid and semi arid climates. If one wishes to find a solution to this problem, bio-drainage cannot be a good alternative to conventional drainage systems.
k (m day−1)
q (mm day−1)
C0 (dS m−1)
C1 (dS m−1)
The followings can be concluded:
The effectiveness of the system is higher where the neighboring strips are narrower; i.e. 25 and 50 m instead of 50 and 100 m for crop strips and plantation strips, respectively. It could therefore be recommended that the designers take the narrowest possible strips into their consideration to such an extent that it does not impair the work of agricultural machinery and does not cause any other limitations.
Maximum initial soil water salinity which can be controlled by bio-drainage is about 5 dS m−1 in a 5 to 10 year time span. In many cases even an initial soil salinity of 3 dS m−1 cannot be tolerated in the rather longer run (for example 20 years).
In most cases salinity of crop strips in bio-drainage is independent from the hydraulic conductivity of the soil. The crop strips could therefore be reclaimed even in heavy soils with low hydraulic conductivity provided that the transmissivity of the soil can satisfy the evapotranspiration of the trees.
Barrier depth does not have an important effect on lowering water table in tree plantation strips. The higher barrier depth, however, has a great influence on lowering the salinization rate of tree plantation strips. A bio-drainage system could not be expected to be considered successful in areas where the barrier depth is too shallow, say less than 4 m in arid and semi arid regions.
Bio-drainage is very sensitive to the amount of applied water. The higher the water applied, the lower is its effectiveness in both water table and salinity control in plantation strips. Choosing a cropping pattern with lower consumptive use and/or increasing water application efficiency may therefore increase the life of the system.
In arid and semi arid areas, sustainability can only be achieved where some sort of salt removal mechanism is included in the system. These could be foliage harvest and scraping salts accumulated on the soil surface in plantation strips. It seems that in some cases bio-drainage can be complemented by traditional drainage systems as Heuperman 2000 and Heuperman et al. 2002 suggests. This is, of course, subject to further research.
The application of this system is more feasible in regions with cheap land and high water price. The applicability of bio-drainage could be higher in areas where the water quality is relatively better and its price is lower while, at the same time, the land is relatively expensive. There is therefore usually a paradox between technical and economic points of view.
The authors would like to sincerely express their gratitude to R. J. Oosterbaan, from International Land Reclamation Institute (ILRI), Wageningen, The Netherlands for his generous assistance in modeling.