ASSESSING THE IMPACT OF BIOENGINEERING TECHNIQUES ON

STREAMBANK EROSION CONTROL

Arnaldo T. Coelho*, Terezinha C. B. Galvão 1, Gustavo B. Menezes 2, Ecidinéia P. S.

Mendonça3, Ênio M. Brandão Fonseca 4

*Ingá Engenharia e Consultoria Ltda.

Rua Santa Fé 100/cj. 304, Sion, Belo Horizonte/MG, Brasil
e-mail: arnaldo@ingaengenharia.com.br, webpage: www.ingaengenharia.com.br

Keywords: Soil bioengineering, erosion control, streambank erosion control, shoreline

protection

Abstract. This paper presents results of bioengineering erosion control techniques

performance from experimental sites located at margins of reservoir lakes, at tropical
conditions. Three experimental sites were located on the margins of the water reservoir at
Volta Grande Hydroelectric Power Plant in Brazil. This reservoir lake occupies an area of 220
km2 and is located in the border of the states Minas Gerais and São Paulo. Twenty-seven
experimental land parcels comprising eleven types of bioengineering techniques were placed
on three experimental sites- Pier, Baia, and Miguelopolis. The total study area covered about
7,500 m2, length of 500 m, and corresponded to the most eroded sites around the reservoir
lake. This paper focused only on the results for the Baia study area. The implemented
bioengineering techniques were as follows: Straw logs and Coir logs (biologs); three types of
metallic gabions (coated with polymers and PVC) - bags, box, and mattress; Polypropylene
(PP) P550 geotextile; PP geotextile C350; Sintemax geotextile; MacMat geotextile; and wood
crib wall.
Laboratorial, field testing and measurements included geotechnical, fertility, turbidity, wind
velocity, wave height, stratigraphy, and in situ permeability studies.
Geotechnical results indicate that the streambank soil is erosion prone. Annual average wind
velocity ranged from 5- 12 km/h, with maximum value of 20 km/h. On the months of March,
May and September, the wind velocity presented large variations with ranges of 6-18km/h, 2-
14 km/h, and 3-17 km/h, respectively. The maximum wave height during 3 years was 60 cm for
the Baia area. This implies that the tractive force of the water varies along the year, the flow
is non-uniform, and flow conditions cannot be considered steady-state. Also, wind velocities
above 5 km/h generated wave heights which varied with power of two with wind velocity.
Collected data from all treatment land parcels over a period of three years were summarized
in a Performance Matrix, which was developed for this project. Turbidity results could not by
themselves provide information on soil loss. Thus, to gather more data on the sediment
accumulated in front of each treatment, differential bathymetry was performed and plotted
using ArcGIS map algebra. Under the studied conditions, the best bioengineering techniques

1
Spelman College, USA
2
California State University, USA
3
Escola de Engenharia da Universidade Federal de Minas Gerais
4
Companhia Energética de Minas Gerais - CEMIG
1
 Arnaldo T. Coelho, Terezinha C. B. Galvão, Gustavo B. Menezes

were MacMat geotextile associated with bag Gabion, and geotextil C350, whereas vegetated
coirs (biologs) and vegetated crib walls had the worst performance. Also, the use of sole
vegetation was not an efficient method for controlling erosion. This experimental study is
relevant because large scale experiments are difficult to implement, to monitor and even to
find in literature. Results may help to shed light on erosion control processes at reservoir
margins, which are very significant to various regions of the world, including Brazil.

1 INTRODUCTION
Traditional engineering practices to control streambank erosion include mainly hard -
armoring techniques such as rock gabions, stone riprap, concrete pavement, among others.
In the 1970s, increased environmental awareness associated with the increased numbe r of
failures of these traditional engineering methods generated the need for alternative
methods. One specific situation that almost always guarantees a failure is the use of hard -
armoring only around the bridge abutments and streambanks, upstream and downstream of
a bridge1. The interest of providing alternative solutions to traditional methods started to
rise, and more environmentally-conscious approaches such as the one provided by
bioengineering techniques became popular, and their benefits and advantages were
gradually re-examined 2. The use of bioengineering techniques is not new but has existed
for centuries. It consists of combining biological, mechanical, and ecological features to
stabilize soil and control erosion 3, 4. However, when applied to streambank erosion control,
it becomes more complex and involves principles of geotechnical and hydraulic
engineering, fluvial geomorphology, ecology, botany, and landscape for soil stabilization.
Three authors have been widely recognized for their foundational work on
bioengineering stabilization research and practice: (i) Hugo Schiechtl 5 (266 citations) &
Schiechtl & Stern 6 (122 citations); (ii) Gray & Sotir 7 (896 citations), Gray & Leiser 8 (646
citations); (iii) Coppin & Richard 9 (566 citations); Greenway 10 (620 citations). Also, the
role of the Department of Agriculture 11 and Corps of Engineers 12 has been recognized.
Almost all the literature produced in the USA after 1980 has been based on the foundation
built by Schiechtl in 1980 1.
In general, streambank and slope stabilization can be achieved by the following: (i)
reinforcing the soil with tensile fibers of the root mass; (ii) increasing shear strength by
reducing pore-pressures through transpiration; and/or (iii) anchoring the slope through deep
root penetration into more stable strata 7, 2, 3.
The role of vegetation is to reduce superficial erosion by intercepting the raindrops,
slowing surface runoff, preventing soil particle detachment from the soil matrix and
increasing water infiltration 4, 7, 9, 10, 13, 14, 15 .
Roots and root architecture can reinforce soil and increase slope stability due to the added
apparent cohesion given by roots 4, 7, 22. The apparent root cohesion is a product of the root
cross-section area per unit area of soil and root tensile strength 18,19. Finer roots are stronger
and provide more resistance than thicker roots 17. A great number of research has been
produced on use of vegetation to control erosional processes 4, 7, 13, 14, 15, 16, 17.
The soil reinforcement provided by roots was derived from a modified Coulomb’s Law shear
strength equation, as an added increment of shear strength, ΔS 20:

𝑆 = 𝑐 + 𝛥𝑆 + 𝜎𝑁 𝑡𝑎𝑛𝜑 (1)
𝛥𝑆 = 𝑇𝑆 (sin 𝛳 + cos 𝛳 tan 𝜑) (𝐴𝑟 /𝐴) (2)

Where, ΔS is the root additional shear resistance; φ is the internal friction angle, σN is the
normal pressure, Ar/A is the ratio of root cross-section area per unit area soil of the shear

2
 Arnaldo T. Coelho, Terezinha C. B. Galvão, Gustavo B. Menezes

plane, TS is the tensile strength of the roots, and  is the angle of the shear plane.

In the last three decades, the use of bioengineering techniques became widespread due to
its intrinsic advantages: low implementation cost, low greenhouse gases footprint,
preservation of local biodiversity and habitats, and better landscape aesthetics, among
others 4, 7, 9, 21, 22, 23.
Despite the great achievements in understanding soil erosion processes, studies are still
lacking on better management practices, and the main erosional processes at water reservoir
margins.
In this study, a wide range of bioengineering erosion control techniques were
implemented/monitored at the margins of water reservoir aiming to identify the main
erosional processes and better management practices that take place. These studies are
relevant to Brazil, as hydroelectrical power plants (UEH) supplies the country with about
92% of its power needs and the number of water reservoirs associated with UEH and
subjected to erosional processes are also big, with a total perimeter of about 6,000 km.

2 METHODOLOGY
2.1 Site Description
This paper focuses on the Baia (3,524 m2) study area located on the reservoir margins at
Volta Grande hydroelectric power plant, Brazil 4. The native vegetation of the area is
mainly composed of Caryocar brasiliense (pequi), Aspidosperma tomentosum (perobeira
do campo), Stryphnodendron adstringens (barbatimão), Copaifera langsdorffii (copaíba do
cabloco), and Tabebuia ochracea (ipê do cerrado). Starting in the 1940s, native vegetation
has been replaced with planted pastures of Brachiaria decumbens and Hyparrenia rufa. In
the 1990s, Pennisetum purpureum was planted in the area as an attempt to stabilize lake
shores and mitigate the erosive effect of wind waves on the reservoir streambanks.
However, this approach was not effective, as the banks kept retreating at a rate of about
1m/year.
The climate of the study area is classified as Cwa according to Köppen, which means
tropical savannah, moist and hot, with a dry season lasting from April to September. 80% of
the rainy season lasts from October and March, and the mean annual precipitation is 1400 mm.
The mean annual temperature is 23.5oC, with minimum temperature around 16.5oC and
maximum temperature around 29.2oC 4. The streambank soil profile is characterized by
homogeneous red latosol 24, also known as Oxisol, according to US Soil Taxonomy25.

2.2 Bioengineering Techniques

Vegetation in combination with a wide range of inert elements (bioengineering techniques)

was implemented in the study area. Bioengineering techniques investigated included rolled
erosion products (RECPs) made of synthetic (geotextile) or natural materials (straw or
coconut fiber) implemented either alone or combined with armored engineering erosion
control techniques (three types of gabion, rip rap and crib wall) in the three study areas. More
specifically, treatment techniques were as follows: (i) rolled erosion product - C350 –
produced by North American Green (NAG), USA; (ii) rolled erosion product -P550 –
NAG/USA; (iii) crib wall; (iv) rip rap; (v) rip rap + vegetated bag Gabion; (vi) vegetated
biologs filled with straw; (vii) vegetated biologs (filled with straw) + Sintemax geotextile
from Deflor Bioengineering/Brazil; (viii) vegetated MCMat (Maccaferri/Brazil) + bag
Gabion; (xix) vegetated biologs filled with coconut fiber (coir logs); (x) vegetated biologs
with coconut fiber + Sintemax; (xi) vegetated Gabion (4m); (xii) vegetated mattress

3
 Arnaldo T. Coelho, Terezinha C. B. Galvão, Gustavo B. Menezes

Gabion (5m); (xiii) Sintemax and MacMat.

To enhance plant growth on top of the treatments, a seed blend that had shown successful
growth under tropical conditions, as well as a great trampling resistance was used ([9], [10],
[11], [12]). The seed blend included the following: Brachiaria humidicola, Pannisetum
purpureum, Panicum maximum, Camipim, Mimosa sp., Cajanus cajan, Melia azedarach,
Clitoria racemosa, Schinus mole, Enterolobium contorstisilquim, and Anadenanthara sp.

2.3 Testings

Laboratorial, field testing and measurements included a wide range of geotechnical studies
(Atterberg Limits, granulometry, density, in situ permeability Guelph, shear strength with soil
and soil/roots soil survey), soil fertility, and laboratory turbidity measurements of water
samples collected in front of each treatment, measurements of wind velocity, and wave height.
The last three were done twice a month for three years.
A survey included topographic measurements of bathymetry and Ground Penetrating Radar
(GPR). Topographic measurements of bathymetry were made twice – before treatment
implementation and 3 years later. Throughout 3 years, two field inspections per year were made
in all the sites to observe and record the integrity of the treatments, maintenance need,
landscape visual aesthetics observations, including the development of erosional spots, and the
toe integrity. Results of monitoring the land parcels were recorded in a Performance Matrix.

3 RESULTS AND DISCUSSION

The below findings are focused only on the study Baia area, which encompasses 11
treatment land parcels, occupying an area of 3,524 m2. Based on results from Atterberg
Limits, soil granulometry, and in situ Guelph permeability testings, soils were classified as
erosion prone (sandy-clayey silt with silt =54%, sand = 26%, and clay = 20%), non-plastic
with IP = 11.9% and permeability Guelph in the range of 10-4 cm/s. Fertility studies showed
that the macro nutrients P, K, and N were low, but all other remaining soil macronutrients
(Ca, Mg, Al, H+Al, organic matter) were on normal range. Soil pH was 6.6, slightly acidic.
Soil erodibility was about 0.40 26.
Figures 1 and 2 show the annual wind velocity (km/h) measured at 1.50 m height, as high,
low and average monthly values, and wave height vs wind velocity for the study area,
respectively. These values for wind velocity indicated that the bank was subjected to variable
traction force along the year, and principles of hydraulics applied to channels cannot be
applied to reservoir margins bank stabilization. The average wind velocity values ranged
from 5- 12 km/h (3-8 miles/h). As shown in Figure 2, wind velocities varying from 0-5 km/h
generated no waves. For wind velocities above 5 km/h, the wave height is proportional to the
wind velocity.

4
 Arnaldo T. Coelho, Terezinha C. B. Galvão, Gustavo B. Menezes

Wind Velocity in km/h

(measured at 1.50m height)
20

Wind Velocity (km/h)

15

10

Month

Figure 1 – Annual wind velocity (km/h)

70

60
y = 0,1071x2 + 0,7525x - 2,2473
Wave Height (cm)

50

40

30

20

10

0
0 5 10 15 20 25
Wind Velocity (km/h)

Figure 2 – Wave height (cm) versus wind velocity (km/h)

Figures 3.a shows erosion at the study area before implementation of bioengineering
techniques. The bank retreat was about 1m/year. Figure 3.b shows clearly that the main type
of bank failure is cantilever failure. The main characteristic of cantilever failure is the basal
removal of sediment by seepage. It leaves a characteristic alcove-shaped indentation
(cantilever), in which the upper bank has no support and is vulnerable to failure.
Figures 4 to 6 depict the implementation of the selected bioengineering treatments, and
after the vegetation growth. Figures 4.a presents the implementation of the C350 geotextile.
Figure 4.b shows vegetation growth on top of C350. Figures 5.a and 5.b show the Gabion
mattress just after implementation, and after a few months after the implementation,
respectively. This was also a very effective treatment that lasts until through today with some
maintenance.
Figure 6.a shows vegetation growth on top of the coconut coirs (biologs) and Figure 6.b
presents the vegetation growth on top of the straw coirs (biologs). It is clear that both biologs
provided a protective barrier against erosion, as indicated by the color of the water in front of
them. Unfortunately, this protection did not last long, and the biologs ended up dissolving in

5
 Arnaldo T. Coelho, Terezinha C. B. Galvão, Gustavo B. Menezes

the lake, first the ones filled by straw followed by the ones filled with coconut fiber. Although
the vegetation was kept behind, it is important to keep in mind that biologs will probably not
last more than six months. Also, the vegetated crib walls were not effective, and the logs
detached after several months from the soil matrix and ended up in the middle of the lake.
Table 1 shows the performance matrix developed based on field observation. The higher the
total score, the better the performance of bioengineering technique.

Figure 3.a – Before treatment Figure 3.b - Cantilever failure before treatment

Figure 4.a – C350 and P550 geotextiles Figure 4.b – Vegetation growth on top of C350

Figure 5.a – Gabion Mattress Figure 5.b – Vegetation growth on Gabion

Figure 6.a – Vegetated coconut coirs Figure 6.b – Vegetated straw coirs

6
 Arnaldo T. Coelho, Terezinha C. B. Galvão, Gustavo B. Menezes

Variable Scores Assigned Scores

Treatment B1 B2 B3 B4 B5 B6 B8 B9 B10 B11

High – 0
integrity on
streambank
spots/ toe
Erosive

Intermediate – 1
Low – 2 0 2 3 3 2 1 2 2 0 0
Inexistent – 3

Bare (<30%) – 0
Vegetative

growth

Low vegetative cover (30 to ≤ 50%) – 1

cover

Average vegetative cover (>50-≤ 70%) – 2 0 1 2 2 3 2 3 3 1 1

High vegetative cover ( >70-100%)- 3

Very High (> 300) – 0

/m2 of margin
treatment (R$

to be treated)

High (150-300) – 1
Cost of

Average (50- 150) – 2 2 3 0 0 1 1 2 2 2 3

Low (<50) – 3

Serious damage (> 30% of the total) – 0

Structural
Integrity
after 24
months

Average damages (10-30% of the total) – 1

0 1 3 3 3 0 2 2 0 0
Low level of damage (<10% of the total) – 2
No damage - 3

High (> de 5 times) – 0

maintenance
Need for

Average (3 to 5 times) – 1
Low (1 – 2 times) – 2 0 1 3 3 3 0 0 0 0 0
No need – 3

No integration with local landscape – 0

Integration with local landscape after 12 months – 1
(aesthetics)
Integration
Landscape

Integration with local landscape after 6 months – 2

0 2 1 1 1 1 2 2 0 0
Integration since the startup – 3

Absence of native flora – 0

vegetation
Regrowth

the parcel
of Native

on top of

Presence of 1-3 native species – 1

Presence of 3-5 species – 2 0 2 2 2 2 3 2 2 0 0
Presence of more than 5 native species – 3

TOTAL 2 12 14 14 15 8 13 13 3 4
Table 1 – Performance Matrix of Baia Experimental Land Parcels (3,524 m2)

Legend: B1 – vegetated biologs (filled with straw) + Sintemax geotextile from Deflor Bioengineering/Brazil;
B2 – rip-rap; B3- rip-rip + Gabion mattress; B4- Gabbion mattress + geotextile; B5- MacMat geotextile from
Maccaferri/Brazil + bag Gabion; B6 – Gabion mattress + MacMat geotextile from Maccaferri/Brasil; B8 -
geotextil C350 from North American Green/USA; B9- geotextil P550 from North American Green/USA; B10-
vegetated biologs (filled with straw) + Sintemax geotextile from Deflor Bioengineering/Brazil; B12- Vegetated
Crib Wall.

Figure 7 depicts the differential topography results for each area/treatment obtained using
ArcGIS geoprocessing tools, which indicated positive results in terms of sediment deposition
in front of each treatment.

7
 Arnaldo T. Coelho, Terezinha C. B. Galvão, Gustavo B. Menezes

B11

B10

Legend
Legenda B9

D (m)==Final
D(m) – Initial
[Bat.Final Bathymetry
- Bat.Inicial]
-0.55 - -0.3
B8
-0.30 - -0.15

-0.15 - -0.05
B7
-0.05 - 0.05

0.05 - 0.15 B6

0.15 - 0.3

0.30 - 0.4 B5

0.40 - 0.55

0.55 - 0.75 0 12.5 25 meters

Metros
B4

Figure 7– Differential bathymetry for the study area of Baia, in 2007

Table 2 summarizes the results from Performance Matrix and Differential Bathymetry
indicating which bioengineering performed the best.

Table 2 – Performance Matrix and Differential Bathymetry Results

Best Treatments
Performance Matrix B5 followed by B3=B4, B8, and B9
Differential Bathymetry B4, B5, and B8

By superimposing these results, we can conclude that for the studied site conditions,
treatements B5 and B8 performed the best – both are armored techniques (B5 - MacMat
geotextile from Maccaferri/Brazil + bag Gabion, and B8 – geotextil C350 from North American
Green/USA).

4 CURRENT WORK (2016 – 2021)

An ongoing study covering about 15,000 m 2 has been implemented to further study
erosional processes on streambank and slope areas. All treatment parcels were implemented
in 2016 and this study will continue until 2021. The streambank experimental parcels are
located at margins of the reservoir lake of the hydroelectric power plant Porto Colombia
and selected areas around the Volta Grande Hydroelectric power plant reservoir lake at the
Rio Grande (Grande River). The reservoirs are located along the border of Minas Gerais
and São Paulo States in Southwest Brazil. The areas were selected for having similar

8
 Arnaldo T. Coelho, Terezinha C. B. Galvão, Gustavo B. Menezes

characteristics to the previous study experimental areas and also for being located within
areas exposed to the highest wind wave intensities for the region.
Fourteen erosion control techniques were implemented on 42 experimental land parcels and
included new cutting-edge methods, as well as some used in previous study, which were either
successful or needed further validation. They were as follows: RECP (rolled erosion control
product) blanket, made with a bottom metallic geogrid layer, and top layer made mixture of
synthetic and coconut fiber geotextil; RECP W3000 from Tensar/USA; vegetated Brazilian
Blankets ROSV (Retentores Orgânicos de Sedimento Vivos); Syntetic Brazilian Blanket – RSS
(Retentores Sintético de Sedimento) and rip ; Gabion + Mattress Gabion + geogrid Triton
(Tensar/USA); Gabion + Mattress Gabion (Macaferri); geoweb Tensar (USA); crib-wall;
Gabion Mattress Maccaferri; Gabion + Gabion Bag + Geogrid Tensar (USA) + blanket
MacMat R3 (Maccaferri); Gabion + Gabion Mattress + RECP MacMat R3 (Maccaferri);
Blanket P550 (Tensar/USA); A-jack concrete system.

5 CONCLUSIONS
Results presented here for the study area indicated that:
1. Soil fertility studies show that the studied soils have low fertility for K and that all other
essential macro nutrients (P, Ca, Mg, Al, H+Al, organic matter) responsible for plant
growth were in normal range. Soil pH was 6.6, slightly acidic.
2. Geotechnical studies show that the studied soils are non-plastic, and the granulometric
analysis classify the soil as sandy-clayey silt.
3. Guelph field permeability indicates that the studied soils have permeability around 10-4
cm/s and are erosion prone. Soil erodibility measured according to [21] was about 0.4
(average erodibility). The maximum value that soil erodibility can reach in RUSLE
equation is 1.
4. Wind velocity below 5km/h has not generated waves, independent of wind direction.
Maximum wave height measured was 0.60 m, within 3 years of monitoring.
5. Turbidity results could not be analyzed as absolute values, as it is virtually impossible to
distinguish between the turbidity generated by waves impact with the upstream turbidity.
6. Treatments B5 and B8 performed the best – both are armored techniques (B5 - MacMat
geotextile from Maccaferri/Brazil + bag Gabion, and B8 – geotextil C350 from North
American Green/USA), whereas vegetated biologs (independent of the filled materials
used) and crib walls had the worst performance.

6 ACKNOWLEGEMENTS
The authors of this paper acknowledge the funding for this study by R&D Program of the
National Agency of Electric Energy (ANEEL – Agência Nacional de Energia Elétrica) with
support of Companhia Energetic of Minas Gerais - CEMIG GT (CEMIG Geração e
Transmissão) and the support for the current work by Electrobras Furnas.

7 REFERENCES
[1] Li, Ming-Han; Eddleman, K. Biotechnical engineering as an alternative to traditional engineering
methods – A biotechnical streambank stabilization design approach. Landscape and Urban
Planning 60, 225-242. (2002).
[2] Riley, A. L. Restoring Streams and Cities. Island Press. Washington, DC. (1998).
[3] Allen, H. H., Leech, J. R. Bioengineering for Streambank Erosion Control. Environmental Impact
Research Program Technical Report EL-97-8. US Army Corps of Engineers Waterways
Experiment Station. (1997).
[4] Coelho, Arnaldo T. Efeitos da Vegetação Herbácea Associada a Técnicas de Bioengenharia de

9
 Arnaldo T. Coelho, Terezinha C. B. Galvão, Gustavo B. Menezes

Solos no Controle De Erosão em Margens de Áreas de Reservatório da UHE Volta Grande, nos
Municípios de Água Comprida/Mg e Miguelópolis/SP. Ph.D Degree Thesis (in Portuguese).
Engineering School. Belo Horizonte, Brazil. (2007).
[5] Schiechtl, H. M. Bioengineering for Land Reclamation and Conservation. University of Alberta
Press. Edmond. Canada. (1980).
[6] Schiechtl, H. M., Stern, R. Water Bioengineering Techniques for Watercourse, Bank and
Shoreline Protection. Blackwell Scientific Publication. Oxford.
[7] Gray D. H.; Sotir, R.B. Biotechnical and soil bioengineering slope stabilization - A practical
guide for erosion control. New York: John Wiley & Sons. (1996).
[8] Gray, D.H. Leiser, A, T. Biotechnical Slope Protection and Erosion Control. Nostrand Reinhold,
New York
[9] Coppin, N.J. and Richards, I.G. Use of vegetation in civil engineering. Sevenoaks, Kent:
Butterworts. (1990).
[10] Greenway, D. R. Vegetation and Slope Stability. In: M.G. Anderson, K. S. Richards (Eds.). Slope
Stability Geotechnical Engineering and Geomorphology, 187-230. Wiley. New York
[11] United States Army Corps of Engineers -USACE. Shoreline and Channel Erosion Protection:
Overview of Alternatives. WRP Technical Note HS-RS-4.1. Wetlands Research & Technology
Center, Environmental Laboratory, U.S. Army Engineer Research and Development Center,
Wetlands Research Program, Vicksburg. (1998).
[12] United States Department of Agriculture - USDA. In Engineering Field Handbook Part 650
Engineering Field Handbook National Engineering Handbook, Part 650, 210-EFH, (2016)
[13] Coelho, A. T.; de Brito Galvão, T. C. Controle de Erosão em Taludes de Disposição de Rejeitos
Siderúrgicos com Tapete Biodegradável. In: VI Simpósio Nacional de Controle de Erosão.
Presidente Prudente, SP. (in Portuguese). (1998).
[14] Coelho, A. T.; de Brito Galvão, T. C Pereira, A.R. Evaluation of Two Biodegradable Rolled
Erosion Control Products (RECP's) for Stabilizing Steep Slopes Under Tropical Conditions. In:
32nd Annual International Erosion Control Association Expo – International Erosion Control
Association, 2001, Las Vegas. Proceedings of the International Erosion Control Association. Las
Vegas. (2001a).
[15] Coelho, A. T.; de Brito Galvão, T. C; Pereira, A.R. The effects of vegetative cover in the erosion
prevention of a road slope. Environmental Management and Health; v. 12, n. 1. (2001b)
[16] Zhang Chao-Bo; Chen Lin-Hua, Jiang J. Why fine tree roots are stronger than thicker roots: The
role of cellulose and lignin in relation to slope stability. Geomorphology. Vol. 206, 196-202.
(2014)
[17] Ye, Chao; Zhonglu Guo; Zhaoxia Li; Chongfa Cai. The effects of Bahia grass on soil erosion
resistance of Aquults in Subtropical China. Geomorphology 285, 82-93. (2017).
[18] Wu T. H., McKinnell III, W. P. Swanston, D. N. Strength of tree roots and landslides on Prince
of Wales Island. Canadian Geotechnical Journal 16(1), 19-33. (1979).
[19] Wu, T.H.; Macomber, R.T.; Erb, R.T.; Beal, P.E. Study of soil-root interactions. Journal of
Geotechnical Engineering (ASCE) 114 (GT12), p. 1351-1375. (1988).
[20] Waldron, L., 1977. The shear resistance of root-permeated homogeneous and stratified soil. Soil
Science Society of America Journal, 41 (5), 843-849. (1977).
[21] McCullah, J., and D. Gray. National Cooperative Highway Research Program NCHRP Report
544 and Project 24-19. Transportation Research Board, Washington, DC. (2005).
[22] Morgan, J. Soil erosion and conservation. U.K.: Silsoe. (1994).
[23] NRCS – Natural Resources Conservation Service. Practice Standard Streambank and Shoreline
Protection (Ft.) Code 580. https://efotg.sc.egov.usda.gov/references/public/AL/ tg580.pdf.
(2012).
[24] EMBRAPA – Empresa Brasileira de Pesquisa Agropecuária. Mapa de Solos do Brasil. (2011).
[25] United States. Department of Agriculture. Soil Conservation Service - USDA. Soil taxonomy: a
basic system of soil classification for making and interpreting soil surveys (No. 436). US
Department of Agriculture. (1975).
[26] Wischmeier, W. H.; Smith, D.D. 1978. Predicting rainfall erosion losses: a guide to conservation
planning. USDA Agricultural Handbook n. 537, Washington, DC. (1978).

10