A computer program for the prediction of
the salinity of soil moisture, ground
water and drainage water,
the depth of the water table, and the
drain discharge in irrigated agricultural
using different (geo)hydrologic
conditions, varying water management
options, including the use of ground
water for irrigation,
and several cropping rotation schedules.
Design requirements of Saltmod
1. The model should be simple and it should not
require a PhD degree to operate it.
2. The input data should be readily available or
relatively easy to obtain.
3. The model should be able to integrate
agricultural, irrigation and drainage practices.
4. The results can be checked by hand.
5. The data can be imported into spreadsheet
(e.g. Excel) for further analysis.
In the soil profile we recognise 4
1 On top of soil surface
2 Root zone/Evapotranspiration zone
3 Transition zone
Salt and water balances are made for each
Inflow = outflow + change in storage
• Maximum 4 agricultural seasons
• Maximum 3 types of land use per season
• (A, B, U) to be defined by user.
• Example Garmsar, 3 seasons
• Nov-Apr A=wheat+barley 30%
• B=0, U=70% (no irrigation)
• May-Aug A=cotton (20%) B=melon (10%)
• U=70% (no irrigation)
• Sept-Oct A=0, B=0, U=100%
• Together with previous cropping data, the crop rotations must
• Rotation code Kr = 0, 1, 2, 3, 4
0 = no crop rotation (fixed areas e.g. Sugarcane, orchards)
4 = full rotation (no fixed areas)
1 = fallow area is fixed, permanent, other crops have full
In Garmsar use Kr=4
The irrigation water can incorporate re-use of groundwater
and drainage water
The complex capillary rise function is simplified and only
critical depth Dc of water table is used, depending on soil type
Characteristics of drainage systems are introduced in a simple way
using well known drain spacing equations. For details see manual.
Leaching of salts occurs as an exponential depletion curve
Example of partition of fallow land (U) into permanent
fallow (Uc) and temporary fallow (U-Uc) when Kr = 1.
Pooling of percolation/leaching (L) and capillary rise (R)
in the part of the area that is under full rotation
The salinity of the topsoil decreases rapidly (good leaching).
Salinity of transition zone first increases slightly, then decreases.
Salinity of the aquifer shows slow reaction.
Case study Egypt, salinity in time
Salin ity (EC i n dS/m)
0 20 40 60 80 100 120
Cr4(rootzone) Cxa(trans.zone) Cd(drainage)
Saltmod calculates average salinity and an empirical frequency
distribution is used to characterise spatial variation
Case study Egypt
Saltmod was used in Egypt see if drainage systems
could be made less expensive.
However, the leaching efficiency of salts was not
known. It had to be found by trials with the model
Various leaching efficiencies were tried: 20%, 40%,
60%, 80% etc.
The salinity results were compared to actually
measured results and the true efficiency could be
found (next slide).
• The leaching efficiency (FLr) is definitely
greater than 0.6, because smaller values give
results that deviate too much from the
• Flr is definitely smaller than 1 for the same
• The true Flr is about 0.8
• A small error of Flr in the range between 0.7
and 0.8 does not have too much influence on
soil salinity, so OK.
Calibrating groundwater flow
• As the groundwater flow was unknown it had to be found
• Previous reports indicated that no upward seepage
occurs but rather some natural drainage through the
• Therefore trials were made with annual values of natural
drainage Gn = 0.0, 0.07, 0.14, 0.21 and 0.28 m
• The results are shown in the next table.
o Depth of water table:
Dw 1st season (summer) 1.0 – 1.1 m
Dw 2nd season (winter) 1.2 – 1.3 m
o Drain discharge:
Gd 1st season 100 – 150 mm
Gd 2nd season 50 – 100 mm
o Compare with previous slide and conclude that the
natural drainage (Gn) to the underground should
be between 0.70 and 0.21 m/year.
o The value of natural drainage Gn cannot be
determined with great precision due to variation
of data, but it cannot be less than 0.07 and more
than 0.21 m/year
o We will accept the average Gn=0.14 as the true
o Herewith Gn is determined by the model and the
model is calibrated.
Simulating drain depth
Is it justified to use in practice drain
depths of 1.0 m instead of standard
practice 1.4 m so that savings can be
made on installation costs ?
Simulating effects of drain depth
Field measurements showed that most crops have no yield reduction
when the depth of water table is 0.6 m. Why make it deeper?
The previous table shows that with drain
depth = 1.0 m:
• the depth of the water table is OK
• the soil salinity is slightly higher, but OK
• the irrigation efficiency is slightly better
• the irrigation sufficiency is slightly better