1

FLOOD ASSESSMENT

FOR THE EMERGENCY WATER SUPPLY TO RAMAPHOSA SETTLEMENT,

DANNHAUSER LOCAL MUNICIPALITY, AMAJUBA DISTRICT MUNICIPALITY,

KWAZULU-NATAL

Compiled by

Dr Bruce Scott-Shaw & Nick Davis

NatureStamp (Pty) Ltd

Tel 078 399 9139

Email bruce@naturestamp.com

Compiled for

Theo Wicks

SLR Consulting (Africa) (Pty) Ltd

Tel 033 343 5826

Email twicks@slrconsulting.com

April 2019

DRAFT REPORT

2

Table of Contents

Table of Contents ........................................................................................................................................... 2

Tables ................................................................................................................................................................. 3

Figures ................................................................................................................................................................ 3

Annexures ......................................................................................................................................................... 3

Specialist Details & Declaration ................................................................................................................. 4

1. INTRODUCTION ................................................................................................................................... 5

1.1 Project Background and Description of the Activity ........................................................... 5

1.2 Terms of reference ......................................................................................................................... 7

1.3 Gauged versus Ungauged Catchments ................................................................................. 7

2. STUDY SITE ............................................................................................................................................ 8

3. METHODOLOGY ................................................................................................................................. 8

3.1 Site Visit .............................................................................................................................................. 8

3.2 Critical Catchment Delineation and River Reach Analysis ................................................ 9

3.3 Design Flood Determination...................................................................................................... 10

3.4 Flood Line Determination ........................................................................................................... 10

3.5 Flood Line Determination for Minor Channels ...................................................................... 11

4. LIMITATIONS AND ASSUMPTIONS ................................................................................................. 11

5. RESULTS AND DISCUSSION ............................................................................................................. 12

5.1 Desktop Hydrological Assessment ........................................................................................... 12

5.2 Design Rainfall ............................................................................................................................... 14

5.3 Hydraulic Modelling ..................................................................................................................... 14

6. CONCLUSION ................................................................................................................................... 21

7. REFERENCES ...................................................................................................................................... 21

3

Tables

Table 1 Details of Specialist ..................................................................................................................... 4

Table 2 Mean monthly rainfall and temperature observed at Dannhauser (derived from

historical data) ............................................................................................................................ 8

Table 3 Data type and source for the Dannhauser assessment .................................................. 8

Table 4 Comparison of values from some of the rainfall stations that were assessed

during the data analysis ......................................................................................................... 12

Table 5 Comparison between the various one-day design rainfall estimation techniques

available for the study site ..................................................................................................... 14

Table 6 Adopted design peak discharge values (m3.s-1) run through HEC-RAS for each

crossing area .............................................................................................................................. 14

Figures

Figure 1 Locality map of the proposed emergency pipeline ........................................................ 6

Figure 2 General site conditions and structures observed during the site visit .......................... 9

Figure 3 Soil Water Assessment Tool (SWAT) watershed delineation tool for sub-catchment

delineation and stream network creation .......................................................................... 9

Figure 4 Longitudinal profile and channel cross sections developed for a section of the

Mtotwane tribuatry .................................................................................................................. 10

Figure 5 GIS model for flood generation in small channels .......................................................... 11

Figure 6 Long-term annual rainfall at Dannhauser ......................................................................... 12

Figure 7 Terrain model showing the proposed pipeline (yellow), relevant rivers (blue) and

the contributing catchment areas (red) ........................................................................... 13

Figure 8 Steady state analysis of the flood event through the cross sections ......................... 15

Figure 9 Typical bridge crossing (portal culvert) incorporated into the flood model ........... 15

Figure 10 1:100 year flood line for the full extent of the pipeline route ...................................... 16

Figure 11 1:100 year flood extent for the Durncol to Hilltop route ............................................... 17

Figure 12 1:100 year flood extent for the Durncol to Hilltop extended route ........................... 18

Figure 13 1:100 year flood extent for the Skombaren route .......................................................... 19

Figure 14 1:100 year flood extent for the Hattingspruit route ........................................................ 20

Annexures

ANNEXURE A Design Rainfall

ANNEXURE B Rational Method Results

ANNEXURE B SDF Method Results

4

Specialist Details & Declaration

This report has been prepared in accordance with Section 13: General Requirements for

Environmental Assessment Practitioners (EAPs) and Specialists as well as per Appendix 6 of GNR 982 –

Environmental Impact Assessment Regulations and the National Environmental Management Act

(NEMA, No. 107 of 1998 as amended 2017) and Government Notice 704 (GN 704). It has been

prepared independently of influence or prejudice by any parties.

The details of Specialists are as follows –

Table 1 Details of Specialist

Specialist Task Qualification and

accreditation Client Signature

Bruce Scott-Shaw

NatureStamp

Design, GIS

& report

BSc, BSc Hon, MSc, PhD

Hydrology

SLR

Consulting

(Africa) (Pty)

Ltd

Date: 16/04/2019

Nick Davis

Isikhungusethu

Environmental

Services

Design &

GIS

BSc, BSc Hon, MSc

Hydrology

SLR

Consulting

(Africa) (Pty)

Ltd

Date: 16/04/2019

Details of Authors:

Bruce is a hydrologist, whose focus is broadly on hydrological perspectives of land use management

and climate change. He completed his MSc under Prof. Roland Schulze in the School of Bioresources

Engineering and Environmental Hydrology (BEEH) at the University of KwaZulu-Natal, South Africa.

Throughout his university career he has mastered numerous models and tools relating to hydrology,

soil science and GIS. Some of these include ACRU, SWAT, ArcMap, Idrisi, SEBAL, MatLab and

Loggernet. He has some basic programming skills on the Java and CR Basic platforms. He has spent

most of his spare time doing field work for numerous companies and researchers. Bruce has

completed his PhD which focuses on rehabilitation of alien invaded riparian zones and catchments

using indigenous trees. The aim is to select Working for Water (WfW) sites throughout the country and

use micro-meteorological techniques to measure the water use of both the indigenous and alien tree

species in the riparian areas. This research will assist in land rehabilitation and restoration in the highly

sensitive riparian areas. A modelling approach has been incorporated into the research to improve

the spatial resolution of the research and to work as a management tool. Bruce has worked on

numerous projects for the CSIR and Ezemvelo KZN wildlife which has included micrometeorological

work, EIAs and wetland mapping for KZN. Bruce has presented his research around the world, where

most recently he represented South Africa at the Singapore International Water Week on water policy

and implementation.

Nicholas Davis is a hydrologist whose focus is broadly on hydrological perspectives of land use

management, climate change, estuarine and wetland systems. Throughout his studies and

subsequent work at UKZN he has mastered several models and programs such as ACRU, HEC-RAS,

ArcMap, QGIS, Indicators of Hydrologic Alteration software (IHA) and Idrisi. He has moderate VBA

programming skills, basic UNIX and python programming skills.

Page | 5

1. INTRODUCTION

1.1 Project Background and Description of the Activity

The proposed project is a rationalisation of the proposal set out in the Amajuba DM, Umzinyathi DM &

Newcastle LM Regional Bulk Water scheme Pre-feasibility study (RHDHV, 2018). The project takes into account

the need for the supply of Skombaren, Hilltop, Hattingspruit and Ramaphosa in the short to medium term.

The Proposed Phase 1 infrastructure is as follows:

Phase 1-1 Skombaren (WSIG funding):

• 2.5Ml reservoir at Skombaren

• 200 and 355 mm uPVC gravity main 8700 m long from Dannhauser command reservoir to Skombaren

via new Concrete reservoir

Phase 1-2 Hilltop (WSIG funding):

• 450mm NB and 400mm Ductile Iron rising main 19700m long from Durnacol to existing Hilltop reservoirs

• Pumping station at Durnacol (2x 250 kw pumps)

• 5Ml Clear water reservoir at Durnacol

Phase 1-3 Hattingspruit and Ramaphosa (WSIG funding):

• 200mm to 315mm uPVC Gravity main 4200 m long to new hatting spruit break pressure tank

• 200mm to 315mm uPVC Gravity main 6500 m long to existing Hattingspruit clear water reservoirs

• 2.5Ml Break pressure tank / Reservoir

• 110 mm NB uPVC Gravity main 4200m to Ramaphosa settlement

• 200 kl elevated reservoir at Ramaphosa

No infrastructure is required for the abstraction of water for supply to the scheme. All water will be sourced

from the existing Durnacol Water Treatment Works. The centre of the project area is the town of Dannhauser,

which is approximately 25 km northwest of Dundee, 18 km north of Glencoe and 33km south of Newcastle, in

the north west of KwaZulu-Natal. The project falls within the Dannhauser Local Municipality, the water services

authority is the Amajuba District Municipality. Figure 1 below shows the project locality and the areas which

need to be serviced by this project.

As part of the specialist requirements, a flood assessment is required. This would assist the design and

placement of supporting infrastructure at each crossing point.

The coordinates for the proposed pipeline are:

28.0091 S & 30.0784 E

The layout of the proposed pipeline and associated infrastructure can be seen in Figure 1.

Page | 6

Figure 1 Locality map of the proposed emergency pipeline

pg. 7

1.2 Terms of reference

NatureStamp has been appointed to conduct a 1:50 and 1:100 year flood line assessment of any significant

stream or river system that may impact on the proposed bulk water scheme.

The terms of reference are as follows -

i. Hydrological assessment, undertaken by the:

a. Analysis of rainfall data available;

b. Analysis of streamflow data available;

c. Determination of the catchment characteristics;

d. Determination of the Manning’s n-values;

e. Analysis of the river reach network; and

f. Estimation of the design flood.

ii. Hydraulic analysis, illustrated by the:

a. Compilation of the river reach model and flood line using HEC-RAS and HEC-geoRAS;

b. Determination of the flood risk and flood hazard throughout the study site; and

c. Recommendation of mitigation options associated with the hydraulic analysis.

iii. Consolidate results in a report with:

a. Flood line maps; and

b. A final flood line report.

1.3 Gauged versus Ungauged Catchments

Flood hydrology assessments can be limited if the information available is scant. In the Tugela area (which

has been experiencing an ongoing severe drought) most of the smaller tributaries (excluding large rivers) do

not flow all year round as they have done in the past. This can be explained by changes in land use through

intensification and increased areas under crops or commercial forests, an increase in water extraction

(irrigation, dams, industrial needs and human needs), cyclic drought and climate change. Much of the flow

in these rivers is not always accurately recorded by weirs. When a flood hydrology assessment is undertaken,

depending on the data available, either gauged or ungauged catchments can be assessed. Gauged data

are the most accurate approach assuming that the data quality is reliable and over a long period of time. In

the absence of such data, an ungauged catchment is assessed using observed rainfall. This data (assuming

it is of good quality) is used as an input to a rainfall-runoff model. The design flood is determined using a

statistical analysis of the rainfall and the catchment characteristics.

In large catchment areas the antecedent moisture content is important for 1:100 year flood events. If the

catchment is very dry before such an event, dams may fill up first from the flood waters and part of the rainfall

may infiltrate, resulting in a reduced flow through the system, whereas a saturated catchment would result in

a shorter lag time and a larger flow volume in the channel. This can lead to a difference in a simulated flood

using design rainfall (ungauged) and a flood using observed streamflow (gauged). Furthermore, the large

flood events are often poorly recorded in weirs due to poor maintenance and overtopping.

For the study area, no streamflow data was available, as such, a detailed rainfall assessment was undertaken

to determine the design rainfall events.

Page | 8

2. STUDY SITE

The site is located within Quaternary Catchment V31G; falling under the Thukela Water Management Area

(WMA) and the uThungulu waterboard. The site is situated in between the Mtotwane (Class C; moderately

modified, WRC 2011) and the Kalbas (Class B; largely natural, WRC 2011) within the catchment area of the

Buffels River. The Mfushane dam is situated near the proposed route.

Rainfall in the region occurs in the summer months (mostly December to February), with a mean annual

precipitation of 848 mm (observed from rainfall station 033509 W). The reference potential evaporation (ETo)

is approximately 1 815 mm (A-pan equivalent, after Schulze, 2011) and the mean annual actual evaporation

is between 1500 – 1600 mm, which exceeds the annual rainfall. This suggests a high evaporative demand and

a water limited system. Summers are warm to hot and winters are cool. The mean annual temperature is

approximately 21.5 ºC in summer and 11.9 ºC in the winter months (Table 2). The underlying geology of the

site is sedimentary Ecca Shale (Vryheid and dark Volksrust) underlain by coal in some areas. The soils overlain

are sandy-clay-loam ranging from Mispah and Glenrosa to Clovelly form in this particular area.

Table 2 Mean monthly rainfall and temperature observed at Dannhauser (derived from historical data)

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Annual

Mean Rainfall (mm) 150 116 90 39 11 1 4 11 32 71 101 118 848

Mean Temperature

(ºC) 21.9 21.7 20.6 17.7 14.4 11.7 11.9 14.1 17.5 18.5 20.1 21.4 17.6

3. METHODOLOGY

The following methodology was followed in order to meet the objectives as detailed in the terms of reference.

The assessment of these systems considered the following databases where relevant:

Table 3 Data type and source for the Dannhauser assessment

Data Type Year Source/Reference

Aerial Imagery 2013 Surveyor General

1:50 000 Topographical 2011 Surveyor General

5m Contour 2010 Surveyor General

River Shapefile 2011 EKZNW

Geology Shapefile 2011 Durban Geological Sheets/National

Groundwater Archive

Land Cover 2014 EKZNW

Water Registration 2013 WARMS - DWS

*Data will be provided on request

3.1 Site Visit

A site visit was conducted by Bruce Scott-Shaw and Nick Davis of NatureStamp on the 12th of March 2019. A

pre-development condition was assumed. The current condition was assessed as follows -

• The vegetation characteristics of the watercourse were assessed for the determination of the Manning’s

n-values;

• The presence and dimensions of any crossings, such as culverts and bridges, that would act as a barrier

to a flood event and that may be damaged during the occurrence of such an event were noted;

• The overall state of drainage channels, streams and rivers was assessed;

• The slope of the study site as well as evidence of flood damage and erosion around the site were noted;

• The state of existing gauging stations (nearby) was assessed to determine if the structure is accurately

recording streamflow (e.g. evidence of under cutting or damaged features); and

• The elevation at the water level and crossing level in order to verify contour data.

Page | 9

The watercourse systems were not flowing at the time of the site visit, excluding the larger systems. As a result

of low-flow conditions, a river profile analysis was possible using a high accuracy GPS.

Figure 2 General site conditions and structures observed during the site visit

3.2 Critical Catchment Delineation and River Reach Analysis

The critical contributing catchment area was determined for use in both the watershed delineation tool and

HEC-HMS and SWAT models. The sub-catchments were delineated using the 5m contour set as an input. This

was used to create a Digital Elevation Model (DEM) that was then used as an input to the watershed tool

(Figure 3).

Figure 3 Soil Water Assessment Tool (SWAT) watershed delineation tool for sub-catchment delineation and stream network creation

Page | 10

3.3 Design Flood Determination

The peak flows for the 1:10, 1:50 and 1:100 flood events were calculated for the catchments using the rational

method, the SCS-SA model and the Standard Design Flood Method as outlined in the SANRAL Drainage

Manual (2013). The 1:10 and 1:50 year events were included for comparative reasons even though they were

not a required output. The SCS-SA model is a hydrological storm event simulation model suitable ideally for

application on catchments that have a contributing catchment of less than 30 km². The model has been

used widely both internationally and nationally for the estimation of flood peak discharges and volume

(Schulze et al., 1992). The type of surface in the drainage basin is also important. The Rational Method

becomes more accurate as the amount of impervious surface, such as pavements and rooftops, increases.

As a result, the Rational Method is most often used in urban and suburban areas (ODOT Hydraulics Manual,

2014).

3.4 Flood Line Determination

Modelling of the flood lines was undertaken using the U.S. Army Corps of Engineers’ HEC-RAS v5.05

programme, which is commonly used throughout South Africa. Numerous cross sections were created

throughout the contributing area (Figure 4). Ineffective areas/hydraulic structures were digitized and included

in the model. Land use coverage was used to determine the Manning’s n-values in a GIS platform. Each cross

section may have had numerous values on either side of the channel depending on the site characteristics.

Manning’s N-values were obtained from the HEC-RAS Hydraulic Reference Manual (2010) for the channel

areas (a value of between 0.03 and 0.04 was used depending on the presence or absence of rock features

and debris). Design flood values were used as an input for the relevant reaches.

Given the slope of the catchment and the distance to downstream hydrological infrastructure, no inundation

within the study site would occur from external features on the watercourse. As such, Normal Depth was

selected for the reach boundary conditions. The slope of the channel was used as the value for the

backwater calculation of the initial condition. Some inundation structures were included in the cross sections

where there were structures present (Figure 4).

Figure 4 provides an overview of one of the impeding structures along the river. A cross-section shows the

delineated area with unique station variables at each site.

Figure 4 Longitudinal profile and channel cross sections developed for a section of the Mtotwane tribuatry

Page | 11

3.5 Flood Line Determination for Minor Channels

As HEC-RAS and HEC-geoRAS are highly sensitive to the resolution of the terrain data used in the model, small

non-perennial channels such as drainage lines are often not captured within the model. In most cases the

flood output is not required for such channels as the flood generated would be negligible. However, it is good

practice to ensure that all channels or drainage lines are adequately covered. As such, the author has

developed a simple model to generate a flood depth through GIS. The model considers the flood generated

for nearby smaller catchments and applies and area weighted correction. The model generates a flood

height based on this estimation within the existing terrain model. Figure 5 provides a schematic of this model.

Figure 5 GIS model for flood generation in small channels

4. LIMITATIONS AND ASSUMPTIONS

In order to apply generalized and often rigid design methods or techniques to natural, dynamic environments,

a number of assumptions are made. Furthermore, a number of limitations exist when assessing such complex

hydrological systems. The following constraints may have affected this assessment:

• Manning’s n - values (the channels roughness coefficient) was estimated. However, n- values in areas

outside of the study area were estimated using a desktop approach due to the extent of the

catchment.

• 5 meter contour interval data and Digital Elevation Models (DEMs) were used in the design flood

estimation (development of the elevation model). Within a 1 km radius of the site, a detailed

topographical survey was undertaken. Given the desktop flood proposed, this resolution was

considered to be of sufficient accuracy for the flood line determination.

• Given the setting of the site (low flow during the site visit) it was difficult to determine which channels

would be fully active in a flood and which are remnant channels which have since been bypassed.

HEC-geoRAS and HEC-RAS models cannot be used to a very high level of accuracy on smaller non-

perrenial systems as they are usually used on larger catchment areas.

Page | 12

5. RESULTS AND DISCUSSION

A detailed desktop assessment was undertaken for the site. This was the point of departure for the calculation

of design flood volumes. These adopted values were then used in the HEC-RAS and HEC-geoRAS models to

route this flood event through the channel.

5.1 Desktop Hydrological Assessment

A detailed assessment of the rainfall stations and weirs was undertaken for the contributing catchment area.

Rainfall stations were considered based on their proximity to the site, altitude and length/reliability of the data

record. In similar vein, flow gauging stations were considered only if good quality data with a reasonable

record length was available. Table 4 provides an overview of the stations within the contributing catchment

area. No flow gauging stations were of relevance to the small catchment areas.

Table 4 Comparison of values from some of the rainfall stations that were assessed during the data analysis

Station ID Observed

MAP (mm)

Altitude

(m) Reliable Latitude Longitude Data Available Station Name

0370780 W 652 1251 19.7 -28.001 29.934 1882-01---2000-08 Springbok

0370807 W 696 1247 34.7 -27.950 29.951 1882-01---2000-08 Chelmsford Dam

0335032 A 855 1331 25.2 -28.034 30.034 1882-01---2000-08 DBN Nav Colliery

0335091 W 681 1355 24.7 -28.017 30.067 1882-01---2000-08 Dannhauser (POL)

0371150 W 747 1349 37.4 -28.001 30.084 1882-01---2000-08 Try Again-Cambrian

Mine

The long-term mean annual rainfall of the site was 692 mm (Figure 6). However, rainfall in excess of 1200 mm

have been recorded in the area, indicating the variable nature of the rainfall at Dannhauser.

Figure 6 Long-term annual rainfall at Dannhauser

The catchment was assessed through a GIS as described in the methodology. Each crossing point and its

associated contributing catchment area is provided in Figure 7. The terrain model shows that there are only

a few significant crossings as the proposed pipeline follows existing roads and railway lines which were

strategically built along gentle upper plateau or catchment divide areas. This has resulted in very small

catchment areas for the crossing points.

0

200

400

600

800

1000

1200

1400

18

82

18

85

18

88

18

91

18

94

18

97

19

00

19

03

19

06

19

09

19

12

19

15

19

18

19

21

19

24

19

27

19

30

19

33

19

36

19

39

19

42

19

45

19

48

19

51

19

54

19

57

19

60

19

63

19

66

19

69

19

72

19

75

19

78

19

81

19

84

19

87

19

90

19

93

19

96

19

99

Rai

nfa

ll (m

m)

pg. 13

Figure 7 Terrain model showing the proposed pipeline (yellow), relevant rivers (blue) and the contributing catchment areas (red)

Crossing 1 (30.0537; -28.0289)

Contributing Catchment Area 6.17 km2

Crossing 3 (30.0446; -27.9897)

Contributing Catchment Area 3.04 km2

Crossing 2 (30.0362; -27.9726)

Contributing Catchment Area 0.3 km2

Crossing 4 & 5 (30.0784; -28.0091)

Contributing Catchment Area 0.8 km2

Crossing 6 (30.1399; -27.9746)

Contributing Catchment Area 10.85 km2

Crossing 7 (30.1287; -27.9919)

Contributing Catchment Area 3.11 km2

Crossing 8 (30.1022; -28.0297)

Contributing Catchment Area 9.56 km2

Crossing 9 (30.1138; -28.0546)

Contributing Catchment Area 3.49 km2

Crossing 10, 11 & 12 (30.1275; -28.0757)

Contributing Catchment Area 25.13 km2

pg. 14

5.2 Design Rainfall

Design rainfall differs from mean annual rainfall as it is rainfall associated with an events rainfall depth for a

specified storm duration and a recurrence interval (frequency of occurrence). The design rainfall used is

dependent on the method used to determine the peak discharge. The SCS-SA method use 1 day-rainfall for

various return periods while the Rational and SDF Methods use rainfall intensity linked to the catchments Time

of Concentration (Tc) and Storm Duration. The Design Rainfall Estimation (DRE) tool which uses observed

rainfall data has been included for comparison.

The results of the design rainfall analysis are summarised below:

Table 5 Comparison between the various one-day design rainfall estimation techniques available for the study site

Return Period Design Rainfall Depth (mm)

SDF DRE SCS-SA (using DRE) Rational

10 Year Return Period 74.77 94.4 94.4 94.4

50 Year Return Period 115.65 126.8 126.8 126.8

100 Year Return Period 133.26 141.1 141.1 141.1

The runoff results obtained for the 1:10, 1:50 and 1:100 year flood events for the various river reaches are

summarised in Table 5. The populated calculation sheets for the SCS and SDF methods can be seen in

Annexure A & B with the results from the design rainfall estimation tool provided in Annexure C. The high

contrast in values is due to the catchment size limitations of the design approaches. It is expected by the

authors that the estimates from the rational and SDF are over designed. This is likely due to larger catchment

areas and rainfall value that may not be representative of the entire catchment (the area is known for

localised storm events). Furthermore, the lack of vegetation and the presence of eroded channels has

resulted in a much shorter time of concentration than what would have occurred in past decades. The design

values indicate that the larger design events were vastly different between models whereas the smaller more

frequent events were very similar between models. This is likely due to the recommended catchment areas

that these models are designed for. Given the results, the rational model was considered to be the most

appropriate model if design rainfall were to be used.

Table 6 Adopted design peak discharge values (m3.s-1) run through HEC-RAS for each crossing area

Crossing No. Return Period

2 5 10 20 50 100 200

1 37.97488 55.8584 71.45839 91.46017 132.7784 178.0152 196.6872

2 21.36578 31.42757 40.20458 51.45817 74.70499 100.1565 110.662

3 39.87534 58.65384 75.03452 96.0373 139.4233 186.9239 206.5304

4 & 5 40.83049 60.0588 76.83186 98.33772 142.7629 191.4014 211.4775

6 43.002 63.25294 80.91805 103.5677 150.3556 201.5808 222.7247

7 34.5819 50.86757 65.07372 83.28839 120.9149 162.1099 179.1136

8 33.57558 49.38734 63.1801 80.86473 117.3963 157.3925 173.9014

9 26.08477 38.36888 49.08443 62.82357 91.20484 122.2778 135.1035

10, 11 & 12 49.77216 73.21137 93.65764 119.8732 174.0273 233.3173 257.79

5.3 Hydraulic Modelling

Various hydraulic models were produced in HEC-RAS and exported to HEC-geoRAS by importing river

centreline, cross sections, water surfaces and flow data from GIS layers and the hydrologic model. This

allowed for inundation mapping and flood line polygons to be generated. The water surface TIN was

converted to a GRID, and then actual elevation model was subtracted from the water surface grid. The area

with positive results (meaning the water surface is higher than the terrain) illustrated the flood area, whereas

the area with negative results illustrated the dry areas not inundated by the flood. Inundation can be seen at

various locations such as around bends.

Page | 15

The 1:100 year flood lines (Figure 10) indicated only a slight edge of the boundary could be damaged in a

flood.

Figure 8 Steady state analysis of the flood event through the cross sections

Figure 9 Typical bridge crossing (portal culvert) incorporated into the flood model

pg. 16

Figure 10 1:100 year flood line for the full extent of the pipeline route

Page | 17

Figure 11 1:100 year flood extent for the Durncol to Hilltop route

Page | 18

Figure 12 1:100 year flood extent for the Durncol to Hilltop extended route

Page | 19

Figure 13 1:100 year flood extent for the Skombaren route

Page | 20

Figure 14 1:100 year flood extent for the Hattingspruit route

pg. 21

6. CONCLUSION

The results provided indicate that the proposed pipeline would cross 1:100 year flood extents as is the nature

of such linear developments. However, the selected route traverses the ridge, resulting in the catchment area

of each crossing being very small. Although there are numerous inundation areas, the flood risk due to flow

velocity is very small.

7. REFERENCES

1. Drainage Manual, The South African Roads Agency Limited (SANRAL), 6th edition, 2013

2. Drainage Manual, The South African Roads Agency Limited (SANRAL), 5th edition, 2006

3. Lynch, SD. 2003: Development of a Raster Database of Annual, Monthly and Daily Rainfall for Southern

Africa, WRC Report No. 1156/1/03, Water Research Commision, Pretoria, RSA.

4. SCHULZE, RE. (2011) Atlas of Climate Change and the South African Agricultural Sector: A 2010

Perspective. Department of Agriculture, Forestry and Fisheries, Pretoria, RSA. pp 387.

5. SCHULZE, RE. (2012) Climate Change and the South African Water Sector: Where from? Where now?

Where to in future? University of KwaZulu-Natal, Pietermaritzburg Campus, South Africa.

6. US Army Corps of Engineers, HEC-GeoRAS version 4.3.93 for ArcGIS 9.3

7. US Army Corps of Engineers, HEC-RAS version 4.1

8. Visual SCS-SA, R.E. Schulze, E.J. Schmidt and J.C. Smithers, University of Natal

pg. 22

ANNEXURE A Design Rainfall

Design Rainfall in South Africa: Ver 3 (July 2012)

User selection has the following criteria:

Coordinates: Latitude: 28 degrees 0 minutes; Longitude: 30 degreess 3 minutes

Durations requested: 2 h, 1 d

Return Periods requested: 2 yr, 5 yr, 10 yr, 20 yr, 50 yr, 100 yr, 200 yr

Block Size requested: 0 minutes

Data extracted from Daily Rainfall Estimate Database File

The six closest stations are listed

Station Name SAWS Distance Record Latitude Longitude MAP Altitude Duration Return Period (years)

Number (km) (Years) (°) (') (°) (') (mm) (m) (m/h/d) 2 2L 2U 5 5L 5U 10 10L 10U 20 20L 20U 50 50L 50U 100

100L 100U 200 200L 200U

DBN NAV COLLIERY, DU 0335032_A 4.0 31 28 2 30 2 827 1385 1 d 60.0 59.8 60.3 80.2 79.9 80.6 94.1 93.8 94.6 107.9 107.5 108.4 126.4

125.4 127.5 140.7 139.2 142.2 155.5 153.5 157.7

BOARDLANDS, DANNHAUSER 0371029_A 4.0 25 27 59 30 1 753 1307 1 d 53.1 52.9 53.3 71.0 70.6 71.3 83.3 82.9 83.7 95.4 95.0 95.9 111.8

110.9 112.8 124.5 123.2 125.8 137.5 135.7 139.5

TRY AGAIN-CAMBRIAN MINE 0371150_W 4.0 46 27 59 30 5 760 1433 1 d 67.4 67.1 67.7 90.1 89.6 90.4 105.7 105.2 106.2 121.1 120.6 121.7

141.9 140.8 143.2 158.0 156.3 159.6 174.5 172.3 177.0

DOORNKOP 0371237_W 11.5 26 27 56 30 8 736 1354 1 d 58.2 57.9 58.4 77.8 77.4 78.1 91.3 90.9 91.7 104.6 104.2 105.1 122.6

121.6 123.6 136.4 135.0 137.9 150.7 148.8 152.9

CHELMSFORD DAM 0370807_W 12.1 39 27 57 29 57 794 1275 1 d 60.2 60.0 60.5 80.5 80.1 80.8 94.4 94.0 94.9 108.2 107.8 108.7 126.8

125.8 127.9 141.1 139.7 142.6 155.9 153.9 158.2

BALLENGLEICH 0370834_W 15.5 59 27 53 29 58 858 1228 1 d 64.7 64.4 64.9 86.4 86.0 86.8 101.4 101.0 101.9 116.2 115.7 116.8 136.2

135.1 137.4 151.6 150.0 153.2 167.4 165.3 169.9

Gridded values of all points within the specified block

Latitude Longitude MAP Altitude Duration Return Period (years)

(°) (') (°) (') (mm) (m) (m/h/d) 2 2L 2U 5 5L 5U 10 10L 10U 20 20L 20U 50 50L 50U 100 100L 100U 200 200L 200U

28 0 30 3 774 1385 2 h 40.8 32.5 49.1 54.5 43.4 65.6 63.9 51.0 77.0 73.3 58.5 88.3 85.9 68.2 103.9 95.6 75.8 115.8 105.6 83.5 128.4

1 d 60.6 48.0 73.4 81.0 64.1 98.0 95.1 75.3 115.1 109.0 86.3 131.9 127.7 100.7 155.2 142.1 111.8 173.0 157.0 123.2 191.9

Page | 23

ANNEXURE B Rational Method Results

Description of Catchment Dannhauser

River detail Mtotwane Tributary

Calculated by BCSS Date 16/04/2019

Physical characteristics

Size of catchment (A) 6.17 km² Rainfall Region

Longest Watercourse 2.3 km Area Distribution Factors

Average slope (Sav) 0.01 m/m Rural (α) Urban (β) Lakes(

γ)

Dolomite Area (D%) 0 % 1 0 0

Mean Annual Rainfall (MAR) 693 mm

Catchment Characteristics Flat/permeable %

r - look up from Table 3C.3 Sparse grass over fairly rough

surface 0.3

Rural (1) Urban (2)

Surface Slope % Factor Cs Description % Factor C2

Vleis and Pans 10 0.05 0.005 Lawns

Flat Areas 60 0.11 0.066 Sandy, flat (<2%) 0.075 -

Hilly 20 0.2 0.040 Sandy, steep (>7%) 0.175 -

Steep Areas 10 0.3 0.030 Heavy soil, flat (<2%) 0.15 -

Total 100

- 0.141 Heavy soil, steep (>7%) 0.3 -

Permeability % Factor Cp Residential Areas

Very Permeable 5 0.05 0.003 Houses 0.4 -

Permeable 25 0.1 0.025 Flats 0.6 -

Semi-permeable 60 0.2 0.120 Industry

Impermeable 10 0.3 0.030 Light industry 0.65 -

Total 100

- 0.178 Heavy Industry 0.75 -

Vegetation % Factor Cv Business

Thick bush and plantation 0 0.05 - City Centre 0.825 -

Light bush and farm-lands 25 0.15 0.038 Suburban 0.6 -

Grasslands 70 0.25 0.175 Streets 0.825 -

No Vegetation 5 0.3 0.015 Maximum flood 1.00 -

Total 100

- 0.228 Total 0 - 0.000

Time of concentration (Tc) Defined Watercourse Notes:

Overland flow Defined watercourse Pre-development Run-off

Latitude: 28°42'

Tc = Longitude:

32°02'

0.74172915

1.5 Hours 0.7 Hours

Run-off coefficient

Return period (years), T 2 5 10 20 50 100 Max

Run-off coefficient, C1 0.546 0.546 0.546 0.546 0.546 0.546 0.546

(C1 = Cs + Cp + Cv)

Adjusted for dolomitic areas, C1D

0.546 0.546 0.546 0.546 0.546 0.546 0.546

(= C1(1-D%)+C1D%(Σ(Dfactor x Cs%))

Adjustment factor for initial saturation, 0.5 0.55 0.6 0.67 0.83 1 1

Ft

Adjusted run-off coefficient, C1T 0.273 0.3003 0.3276 0.36582 0.45318 0.546 0.546

( = C1D x Ft)

Page | 24

Combined run-off coeffiecient CT

0.273 0.3003 0.3276 0.36582 0.45318 0.546 0.546

(= αC1T + βC2 + γC3)

Rainfall

Return period (years), T 2 5 10 20 50 100 Max

Point Rainfall (mm), PT 60.2 80.5 94.4 108.2 126.8 141.1 155.9

Point Intensity (mm/hour), PiT (=PT/TC) 81.2 108.5 127.3 145.9 171.0 190.2 210.2

Area Reduction Factor (%), ARFT 100 100 100 100 100 100 100

Average Intensity (mm/hour), IT 81.2 108.5 127.3 145.9 171.0 190.2 210.2

(= PiT x ARFT)

Return period (years), T 2 5 10 20 50 100 Max

Peak flow (m³/s), 37.975 55.858 71.458 91.460 132.778 178.01

5 196.68

7

Page | 25

ANNEXURE C SDF Method Results

Description of catchment Dannhauser

River detail Mtotwane Tributary

Calculated by BCSS Date 16 April 2019

Physical characteristics

Size of catchment (A) 6.17 km²

Time of Concentration

(TC)

0.74 hours

Longest watercourse (L) 2.3 km

Average slope (Sav) 0.01 m/m

SDF basin (0)# 26 Time of concentration, t (= 60 TC) 45 minutes

2-year return period rainfall (M) 61 mm Days of thunder per year (R) 17 days/year

TR102 n-day rainfall data

Weather Service station Nqutu Mean annual precipitation (MAP) 760 mm

Weather Service station number 336 283 Coordinates

Duration (days) Return period (years)

2 5 10 20 50 100 200

1 61 84 102 121 150 175 202

2 76 105 128 152 187 217 250

3 84 117 141 168 205 237 272

7 108 151 182 215 263 302 345

Rainfall

Return period (years), T 2 5 10 20 50 100 200

Point precipitation depth (mm) Pt,T 33.88 57.16 74.77 92.38 115.65 133.26 150.87

Area reduction factor (%), ARF (= (90000-12800lnA+9830lnt)0,4)

100% 100% 100% 100% 100% 100% 100%

Average intensity (mm/hour), IT (= Pt,T x ARF / TC) 45.68 77.06 100.80 124.54 155.92 179.66 203.40

Run-off coefficients

Calibration factors C2 (2-year return period) (%) 15 C100 (100-year return period) (%) 50

Return period (years) 2 5 10 20 50 100 200

Return period factors (YT) 0 0.84 1.28 1.64 2.05 2.33 2.58

Run-off coefficient (CT),

0.15 0.28 0.34 0.40 0.46 0.50 0.54

Peak flow (m³/s), QT = 0.278 x CTITA 11.74 36.48 59.13 84.60 122.38 153.96 187.40

Table 3C.7

2 5 10 20 50 100 200

TC < 6 hours Modified Hershfield equation 20.95 35.35 46.24 57.13 71.52 82.41 93.30

6 hours < TC < 24 hours

Linear interpolation between modified Hershfield equation point rainfall and 1-day point rainfall from TR102

- - - - - - -

TC > 24 hours Linear interpolation between n-day point rainfall values from TR103

- - - - - - -