Mr Hamza Briak, Faculty of Sciences and Techniques, Morocco. Global Symposium on Soil Erosion (GSER19), 15 - 17 May 2019 at FAO HQ.

1. Effectiveness Analysis of Agricultural BMPs by SWAT
Model for Appropriate Control of Sediment Yield on
the Kalaya Catchment in the North of Morocco.
Hamza BRIAK
FST – Tangier / INRA – Rabat
Morocco
1

2. 2
Objective
The aim of this research is to evaluate the effects of different
agricultural best management practices (BMPs) on sediments using
Soil and Water Assessment Tool (SWAT) model in the Kalaya river
basin in order to recommend the most appropriate one.

3. 3
Introduction
o Critical environmental, ecological and economic troubles worldwide can
be caused by soil erosion, and has become a challenging issue, menacing
the Mediterranean territories especially in north of Morocco.

4. 4
Introduction
o Critical environmental, ecological and economic troubles worldwide can
be caused by soil erosion, and has become a challenging issue, menacing
the Mediterranean territories especially in north of Morocco.
o In fact, Mediterranean climate is identified by seasonal contrast, where the
soil erodibility can be affected by the dry and warm summer climate, and
where the soil erosion can be affected by the concentration of
precipitation events, particularly in the fall.

5. 5
Introduction
o Critical environmental, ecological and economic troubles worldwide can
be caused by soil erosion, and has become a challenging issue, menacing
the Mediterranean territories especially in north of Morocco.
o In fact, Mediterranean climate is identified by seasonal contrast, where the
soil erodibility can be affected by the dry and warm summer climate, and
where the soil erosion can be affected by the concentration of
precipitation events, particularly in the fall.
o This event may lead to serious degradation of the hill slopes with further
negative damage on natural resources.

6. 6
Introduction
o However, soil loss and sediment discharge to waterbodies can be
prevented and mitigated by best management practices (BMPs). So, to
reduce the soil erosion intensity, it is required to clarify the sources zones
of sediment yield where soil conservation works have to focus on.

7. 7
Introduction
o However, soil loss and sediment discharge to waterbodies can be
prevented and mitigated by best management practices (BMPs). So, to
reduce the soil erosion intensity, it is required to clarify the sources zones
of sediment yield where soil conservation works have to focus on.
o The environmental effectiveness of BMPs in terms of sediment yield
reduction can be assessed by several watershed models. The model
selected for this work is the Soil and Water Assessment Tool (SWAT)
which is one of many models widely used to assess soil erosion risk and
simulate conservation measures efficiency.

8. 8
Study Area
o The research area is the 38 Km2
Kalaya catchment in north of
Morocco, located in the southern
part of Tangier city.
Map of The study area (Kalaya catchment)

9. 9
Study Area
o The research area is the 38 Km2
Kalaya catchment in north of
Morocco, located in the southern
part of Tangier city.
o The study area has a
Mediterranean climate sub-humid
to wet, characterized by wet
winters and dry summers.
Map of The study area (Kalaya catchment)

10. 10
Study Area
o The research area is the 38 Km2
Kalaya catchment in north of
Morocco, located in the southern
part of Tangier city.
o The study area has a
Mediterranean climate sub-humid
to wet, characterized by wet
winters and dry summers.
o The annual average precipitation is
667 mm, while the annual average
temperature is about 18 °C.
Map of The study area (Kalaya catchment)

11. 11
Study Area
o The research area is the 38 Km2
Kalaya catchment in north of
Morocco, located in the southern
part of Tangier city.
o The study area has a
Mediterranean climate sub-humid
to wet, characterized by wet
winters and dry summers.
o The annual average precipitation is
667 mm, while the annual average
temperature is about 18 °C.
o The hydrological monitoring is
ensured by the hydrometric station
located at the outfall Kalaya
(reservoir).
Map of The study area (Kalaya catchment)

12. 12
Modelling Tool
Agro-hydrological
model
Semi-distributed
approach,
physically-based
Continuous-time
model
Simulations:
Hydrology; Erosion;
Nutrients.
Developed by Department of Agriculture,
Agricultural Research Service (USDA-ARS)
o Widely used over the world, Soil and Water Assessment Tool (SWAT) is a
hydrological model to assess sediment, impact of land use, in-stream water
quality, climate change and, water quality and quantity variation.

13. 13
Modelling Tool
Agro-hydrological
model
Semi-distributed
approach,
physically-based
Continuous-time
model
Simulations:
Hydrology; Erosion;
Nutrients.
Developed by Department of Agriculture,
Agricultural Research Service (USDA-ARS)
o SWAT can determine as well as river basin-scale, continuous-time model that
operates on a daily or sub-daily time step, computationally efficient, and able
to estimate long-term yields in large watersheds.

14. 14
Modelling Tool
Agro-hydrological
model
Semi-distributed
approach,
physically-based
Continuous-time
model
Simulations:
Hydrology; Erosion;
Nutrients.
Developed by Department of Agriculture,
Agricultural Research Service (USDA-ARS)
o The model is physically-based, can simulate water in soil and groundwater,
crop growth, erosion, sediments, nutrients and environmental impact of
climate change.

15. 15
Land phase Routing phase
Modelling Tool
o SWAT breaks down the operation of the basin into 2 main components:

16. 16
Land phase Routing phase
Modelling Tool
o SWAT breaks down the operation of the basin into 2 main components:
 The land phase, where it simulates the movement of water, sediments and
nutrients from the emission or production area to the river system.

17. 17
Land phase Routing phase
Modelling Tool
o SWAT breaks down the operation of the basin into 2 main components:
 The land phase, where it simulates the movement of water, sediments and
nutrients from the emission or production area to the river system.
 The routing phase, where it represents the routing of these elements along
the streams to the outlet of the basin.

18. 18
Hydrology Erosion
Water Balance Equation Modified Universal Soil Loss
Equation (MUSLE)
Y = 11.8 * (Q * qp)^0.56 * K * LS * C * P
Where :
o SWt is the soil water content (mm);
o SW0 is the water available to plants
(mm);
o Rday is the precipitation (mm);
o Qsurf is the surface runoff (mm);
o Ea is the evapotranspiration (mm);
o Wseep is the percolation (mm);
o Qgw is the low flow (mm);
o t is the time (days).
Where :
o Y is the sediment yield;
o Q is the runoff volume (m^3);
o qp is the peak flow rate (m^3/s);
o K is the soil erodibility factor;
o LS is the slope length and gradient
factor;
o C is the cover management factor;
o P is the erosion control practice factor.
Modelling Tool

19. 19
Implementation
of the SWAT
Model
Calibration and
validation of
SWAT Model
Simulation of
BMPs
Meteorological
and Hydrological
Data
Topographic and
Hydrographic
Data
Pedological Data
Land Use Data
Flows and
Sediment
Concentrations
Data
Identification of
Sensitivity
Parameters
Monthly Simulation
Cal. (1976 – 1984)
Val. (1985 – 1993)
Evaluation of Model
Performance
Analysis of Results
and Choice of the
Most Appropriate
BMP
Implementing a
suite of BMPs
Model Execution
(1971 – 1975)
Results of the Water
Balance and
Sediments

20. 20
Implementation of BMPs
 Contouring  Strip-cropping
 Terracing

21. 21
Implementation of BMPs
Effect of contouring, terracing, and residue
management on curve number CN (Arabi, et al., 2008).

22. 22
Implementation of BMPs
USLE_P factor values for contouring, strip-cropping, and terracing
(adapted from Wischmeier & Smith, 1978; in Arabi et al., 2008)

23. 23
Observed and Simulated Monthly Stream Flow
for Model Calibration (1976-1984)
Observed and Simulated Monthly Sediment
Concentration for Model Calibration (1976-1984)
Model Calibration
o These figures show that the variation of flow and sediment concentration in the station
is simulated successfully by the model; it represents the monthly peaks marking these
two states.

24. 24
Scatter Plot of Monthly Stream Flow
for Calibration Period (1976-1984)
Scatter Plot of Monthly Sediment Concentration
for Calibration Period (1976-1984)
Model Calibration
o The calibration allowed us to get a good model performance ​​for flow rates with a
coefficient of NSE of the order of 0.76 and PBIAS of the order of −11.80, and also a good
model performance ​​for sediment concentration which NSE of the order of 0.69 and
PBIAS of the order of 7.12.

25. 25
Scatter Plot of Monthly Stream Flow
for Calibration Period (1976-1984)
Scatter Plot of Monthly Sediment Concentration
for Calibration Period (1976-1984)
Model Calibration
o This decreases the doubts associated with this calibration and what's more, it provides
a better estimate of the studied process.

26. 26
Observed and Simulated Monthly Stream Flow
for Model Validation (1985-1993)
Observed and Simulated Monthly Sediment
Concentration for Model Validation (1985-1993)
Model Validation
o Validation of SWAT model was performed over other period of calibration (9 years) by
comparing the flow rates and sediment concentration of measured at flow rates and
sediment concentration simulated in hydrometric station considered.

27. 27
Scatter Plot of Monthly Stream Flow
for Validation Period (1985-1993)
Scatter Plot of Monthly Sediment Concentration
for Validation Period (1985-1993)
Model Validation
o The validation allowed us to obtain a good model performance ​​for flow rates with a
coefficient of NSE of the order of 0.67 and PBIAS of the order of −14.44, and also a good
model performance ​​for sediment concentration which NSE of the order of 0.70 and
PBIAS of the order of 15.51.

28. 28
Scatter Plot of Monthly Stream Flow
for Validation Period (1985-1993)
Scatter Plot of Monthly Sediment Concentration
for Validation Period (1985-1993)
Model Validation
o The good agreement between simulations and observations through the validation
phase also shows the good performance of the model calibration and ability to
represent various climatic situations.

29. 29
Sediment Yield of the Kalaya basin.
Sediment yield
o The quantity of
sediment supplied by
the various units
space of watershed
Kalaya varies between
20 and 120 t/ha/yr,
with an average rate
of around 55 t/ha/yr.

30. 30
Sediment Yield of the Kalaya basin.
Sediment yield
o The highest
sedimentation rates
(particularly at sub-
basins 4, 17, 21, 24, 26
and 27) are located in
agricultural and arable
lands on steep slopes.

31. 31
Sediment Yield of the Kalaya basin.
Sediment yield
o This is probably due
to the influence of
tillage practices on
soil loss because they
increase soil erosion
rate and sediment
losses.

32. 32
Sediment yield after contouring simulations
at the Kalaya catchment
BMPs
 Contouring
o The sediment yield in
Kalaya Basin varies
between 20 and 140
t/ha/yr, with an
average rate of about
70 t/ha/yr.

33. 33
Sediment yield after contouring simulations
at the Kalaya catchment
BMPs
 Contouring
o The sediment yield in
Kalaya Basin varies
between 20 and 140
t/ha/yr, with an
average rate of about
70 t/ha/yr.
o The sub-basins most
affected by erosion
are 4, 21, 26 and 27.

34. 34
Sediment yield after strip-cropping simulations
at the Kalaya catchment
BMPs
 Strip-cropping
o The sediment
production estimated
by the SWAT model
varies between 15 and
80 t/ha/yr, with an
average rate of about
49 t/ha/yr.

35. 35
Sediment yield after strip-cropping simulations
at the Kalaya catchment
BMPs
 Strip-cropping
o The sediment
production estimated
by the SWAT model
varies between 15 and
80 t/ha/yr, with an
average rate of about
49 t/ha/yr.
o The most degraded
sub-basins are 4, 12,
17, 21, 24, 26 and 27.

36. 36
Sediment yield after terracing simulations
at the Kalaya catchment
BMPs
 Terracing
o The amount of
sediment provided by
the different units
varies between 10
and 70 t/ha/yr, with
an average rate of
about 38 t/ha/yr.

37. 37
Sediment yield after terracing simulations
at the Kalaya catchment
BMPs
 Terracing
o The amount of
sediment provided by
the different units
varies between 10
and 70 t/ha/yr, with
an average rate of
about 38 t/ha/yr.
o The highest sediment
levels are observed
in sub-basins 17, 21
and 24.

38. 38
Comparison between
agricultural BMPs
Average annual sediment yield simulated for each treatment
o effective measures to reduce sediment
losses at the watershed level are
organized according to their
effectiveness, and these are terracing
(28% reduction and the value is
15t/ha/yr) followed by strip-cropping
(9% reduction and the value is
5t/ha/yr). On the other hand,
measurements performed by the
contouring are inappropriate for the
study area because they have
contributed to increasing the soil
erosion (more than 31% of losses and
the value is 17t/ha/yr more than
existing conditions).

39. 39
Comparison between
agricultural BMPs
Average annual sediment yield simulated for each treatment
o The mean annual values of sediment
yields obtained for scenarios with and
without BMPs were compared to
assess the effectiveness of BMPs.

40. 40
Comparison between
agricultural BMPs
Average annual sediment yield simulated for each treatment
o The mean annual values of sediment
yields obtained for scenarios with and
without BMPs were compared to
assess the effectiveness of BMPs.
o Among all other practices, terracing
were the most effective BMPs for
reducing sediments.

41. 41
Comparison between
agricultural BMPs
Average annual sediment yield simulated for each treatment
o The mean annual values of sediment
yields obtained for scenarios with and
without BMPs were compared to
assess the effectiveness of BMPs.
o Among all other practices, terracing
were the most effective BMPs for
reducing sediments.
o This indicates that the use of terracing
on agricultural land can potentially
make improvements marked the
control and limitation of soil erosion
which is consistent with previous
research.

42. 42
Conclusion
o The model used was successfully calibrated and validated on the Kalaya river basin.

43. 43
Conclusion
o The model used was successfully calibrated and validated on the Kalaya river basin.
o This system will integrate analysis data, a geographic information system (GIS) and
modeling tools.

44. 44
Conclusion
o The model used was successfully calibrated and validated on the Kalaya river basin.
o This system will integrate analysis data, a geographic information system (GIS) and
modeling tools.
o The system and model developed in the Moroccan context will give a clear picture of
areas where erosion risk is most likely in the study area. They will evaluate the impact
of different corrective methods on this risk and will allow to select the practices best
adapted to each zone to solve the problems.

45. 45
Conclusion
o The model used was successfully calibrated and validated on the Kalaya river basin.
o This system will integrate analysis data, a geographic information system (GIS) and
modeling tools.
o The system and model developed in the Moroccan context will give a clear picture of
areas where erosion risk is most likely in the study area. They will evaluate the impact
of different corrective methods on this risk and will allow to select the practices best
adapted to each zone to solve the problems.
o Different soil conservation scenarios were simulated and tested with respect to
sediment yield using SWAT model.

46. 46
Conclusion
o The model used was successfully calibrated and validated on the Kalaya river basin.
o This system will integrate analysis data, a geographic information system (GIS) and
modeling tools.
o The system and model developed in the Moroccan context will give a clear picture of
areas where erosion risk is most likely in the study area. They will evaluate the impact
of different corrective methods on this risk and will allow to select the practices best
adapted to each zone to solve the problems.
o Different soil conservation scenarios were simulated and tested with respect to
sediment yield using SWAT model.
o The simulation of agricultural best management practices (BMPs) by the SWAT model
provides to have an idea on the most effective practice for reducing sediment yield in
the watershed.

47. 47
Conclusion
o The model used was successfully calibrated and validated on the Kalaya river basin.
o This system will integrate analysis data, a geographic information system (GIS) and
modeling tools.
o The system and model developed in the Moroccan context will give a clear picture of
areas where erosion risk is most likely in the study area. They will evaluate the impact
of different corrective methods on this risk and will allow to select the practices best
adapted to each zone to solve the problems.
o Different soil conservation scenarios were simulated and tested with respect to
sediment yield using SWAT model.
o The simulation of agricultural best management practices (BMPs) by the SWAT model
provides to have an idea on the most effective practice for reducing sediment yield in
the watershed.
o This work will then provide useful information for targeted management and help
interested and involved stakeholders in water and soil conservation activities to
choose the most appropriate practice for the study area.

48. 48