D422Lot1.SMHI.5.1.1B: Detailed workflows of each case-study on … · 2019-02-18 · the streams with raw sewage. The situation is expected to aggravate due to climate change, which

C3S_422_Lot1_SMHI – D5.1.1B |

Copernicus Climate Change Service

D422Lot1.SMHI.5.1.1B:

Detailed workflows of each case-study on how to use the CDS for CII

production and climate adaptation

Full Technical Report:

Potential changes in the water quality

of the rivers in the uMngeni catchment

under a future climate

Michele Toucher, Sean Thornton-Dibb and Mark Horan

University of KwaZulu-Natal

REF.: C3S_422_Lot1_SMHI D5.1.1B

C3S_422_Lot1_SMHI – D5.1.1B |


Summary

The water utility, Umgeni Water, is responsible for supplying potable water to consumers within the uMngeni catchment and in the greater Durban-

Pietermaritzburg area in South Africa. The potential impact of a changing climate on both quality and quantity of the water resource is the 3rd highest risk faced by Umgeni Water. This case study focused on improving the understanding of the

potential changes in water quality in the uMngeni catchment under a future climate. To achieve this the changes in the air temperatures and precipitation

had to be assessed, and these used as input to a hydrological model to determine the potential changes in accumulated streamflows in the water

management units in the catchment by 2040. The Climate Impact Indicators used to generate the future climate scenarios were provided within the C3S_422_Lot1_SMHI contract.

Although uncertainty still remains in the changes in precipitation and, thus

accumulated streamflows, the increases in air temperature are clear. This will have impacts on the water quality of the catchment through an increase in water temperature and an increase in evaporation, potentially resulting in an increased

spread and reproduction rate of water borne diseases and pathogens. Subsequently, this means an increased risk to an already vulnerable set of

communities living close to or reliant on the water courses. This project through the production of CIIs relevant to water quality assessment is addressing these issues and allowing for adaptation planning being undertaken.


C3S_422_Lot1_SMHI – D5.1.1B |

Contents 1- Case study description ..................................................................................... 4

1.1 Issue to be addressed ................................................................................... 4

1.2 Decision support to client ............................................................................... 4

1.3 Temporal and spatial scale ............................................................................. 5

1.4 Knowledge brokering ..................................................................................... 5

2- Potential adaptation measures .......................................................................... 5

2.1 Lessons learnt .............................................................................................. 5

2.2 Importance and relevance of adaptation .......................................................... 6

2.3 Pros and cons or cost-benefit analysis of climate adaptation ............................... 6

2.4 Policy aspects ............................................................................................... 6

3- Contact .......................................................................................................... 6

3.1 Purveyors .................................................................................................... 6

3.2 Clients/users ................................................................................................ 6

4- Data production and results .............................................................................. 7

4.1 Step 1: Data Collection and Model Configuration ............................................... 7

4.2 Step 2: Defining scenarios and CIIs................................................................11

4.3 Step 3: Scenario Modelling and Analysis .........................................................13

4.4 Step 4: Development of Showcase .................................................................21

4.5 Step 5: Dissemination and Communication .....................................................24

5- Conclusion of full technical report .....................................................................24

References ..........................................................................................................25


C3S_422_Lot1_SMHI – D5.1.1B | 4

1- Case study description

1.1 Issue to be addressed The uMngeni catchment in South Africa is highly diverse in climate and land use. The catchment has two large urban areas, Durban and Pietermaritzburg, as well

as several smaller urban areas which are diverse in nature from wealthy, upper class residential suburbs to informal settlement areas which lack access to basic

services. The uMngeni catchment supports a range of intensive commercial agricultural practices including sugarcane, dairy and commercial forestry.

Scattered between these commercial agricultural areas are subsistence farming areas and poor rural communities who lack access to services. The water-related infrastructure which exists in the catchment is under severe strain. Coupled with

this is the need to expand the infrastructure to those areas currently under-serviced (e.g. stand pipes; pit latrines) or lacking accessing to basic services.

The agricultural activities in the catchment, lack of access to sanitation in some areas and aging sanitation infrastructure in others as well as the pressures on the solid waste management have degraded the water quality of the catchment.

The water utility - Umgeni Water - is responsible for supplying drinking water to those areas with access to piped potable water. This water is sourced from four

large reservoirs in the catchment which are at risk of becoming eutrophic. Further to this, those communities without access to piped water are dependent on the river system. The risk of eutrophication of the large reservoirs is due to

agricultural activities, aging infrastructure and a large component of society who have no access to basic sanitation. These activities are also resulting in a

significant decline in the river health and quality on which many depend. Public health and safety are affected when bacteria counts exceed 400 per 100

ml according to South African guidelines; the actual concentrations in the uMngeni river systems have reached up to 1 000 000. A changing climate into

the future could exacerbate this problem.

1.2 Decision support to client The climate impact indicators (CIIs) produced through this project is assisting

Umgeni Water to better understand how the water availability, nutrient and pathogen loadings will change temporally and spatially through the uMngeni

catchment under future climate scenarios. With this understanding, critical areas of the river system which have good water

quality can be identified by Umgeni Water. Actions can then be taken to preserve and protect these areas. Further to this, the understanding can identify hotspot

areas of poor water quality where interventions are needed. These interventions could include rehabilitation of degraded landscapes to restore ecosystem services.

At a more local scale, the irrigation boards and conservancies in the catchment,

particularly in the upper areas of the uMngeni catchment can use the CIIs to better understand how the water quantity and quality in their areas may change under a future climate. This could facilitate the identification of changes in

agricultural water use related practices that are needed to preserve or improve


C3S_422_Lot1_SMHI – D5.1.1B | 5

water quality, such as improved slurry water management and irrigation

scheduling. For the conservancies who champion maintaining the areas of pristine and natural landscapes, it could assist in educating about and managing

land degradation to prevent the currently unimpacted areas becoming areas of concern in the future.

1.3 Temporal and spatial scale The hydrological modelling was undertaken for the greater uMngeni catchment (4 349.4 km2) at a hydrological homogenous response unit scale with results summarized to the thirteen water management units that have been delineated

in the catchment.

The period 1960 – 1999 will be used to produce the historical (benchmark) CIIs as high quality data is available for this period. Future climate scenarios to 2040 were considered.

1.4 Knowledge brokering The Centre for Water Resources Research (CWRR) has a close and ongoing relationship with the client, Umgeni Water. Umgeni Water support an endowed

research Chair i.e. the Umgeni Water Chair of Water Resources Management, who is housed in the CWRR at the University of KwaZulu-Natal. The project team

meets monthly with the Umgeni Water Chair to discuss the data produced and how they are communicated through the interactive Atlas. The Umgeni Water Chair reports these activities at the Umgeni Water Internal Research Direction

meetings and in quarterly reports. Using the Chair as a facilitator for communication ensures that the CIIs are in alignment with the needs of the

client and feedback reaches the correct decision makers in Umgeni Water. The CDS was promoted as a source of climate data, and the interactive atlas demonstrated to highlight the areas that are a) already under stress with regards

to water quality and quantity b) will likely become stressed c) will be stressed but can be contained perhaps with intervention now.

Further to this, the irrigation boards and conservancies in the catchment are considered clients, to whom the interactive climate atlas is being presented and

promoted.

2- Potential adaptation measures

2.1 Lessons learnt The uMngeni catchment has been under drought conditions for the past two

years. During this time the focus has been on the provision of water for today and the month to come, with little attention being paid to further ideas and issues. With the recent good rains and the drought restrictions being lifted, the

focus can shift from coping to planning and adaptation for the future. The lesson learnt from this is that communication with decision makers around long-term

planning and adaptation under crisis conditions is problematic. Provision of long-term information related to planning and adaptation needs to happen during non-crisis times.


C3S_422_Lot1_SMHI – D5.1.1B | 6

2.2 Importance and relevance of adaptation To date there is very limited information available on the impacts of future climate on water quality for South African catchments, despite the recognition

and acceptance of the significant potential impact on water quality. This project through the production of CIIs relevant to water quality assessment for the

uMngeni catchment is partially addressing this and allowing for adaptation planning to be undertaken. This adaptation plan is envisioned to be through the identification of critical areas which need to preserve, to ensure good water

quality, and identification of hotspot areas of poor water quality where immediate intervention is required. Without intervention, the eutrophication of the large

reservoirs and decline of the health of the rivers in the catchment will continue and possibly be exacerbated due to a changing climate, placing the water supply of the communities at risk and increasing the water treatment costs.

2.3 Pros and cons or cost-benefit analysis of climate adaptation Without intervention and improvement of the water quality in the uMngeni catchment, the pressure placed on the limited water treatment facilities will

continue to escalate as will the cost of treating the water. Declines in river health will cause the migration of emerging farmers to the urban areas where

infrastructure is already under pressure. Water-borne diseases will likely increase, overloading an already stretched primary health care system. Service

delivery protests related to water provisioning and sanitation have already manifested in the catchment. Any further decline in the resource may exacerbate the already volatile situation.

2.4 Policy aspects By identifying hotspot areas of concern, those economic land use activities upstream can be monitored and regulated. Further to this, the linking of nutrient

loading and concentrations to activities on the landscape can assist in land use planning.

3- Contact

3.1 Purveyors Michele Warburton-Toucher, Graham Jewitt, David Clark, Mark Horan and Sean

Thornton-Dibb. Centre for Water Resources Research, University of KwaZulu-Natal (UKZN),

Scottsville, South Africa

3.2 Clients/users Umgeni Water, KwaZulu-Natal, South Africa

http://cwrr.ukzn.ac.za/homepage

http://www.umgeni.co.za/


C3S_422_Lot1_SMHI – D5.1.1B | 7

4- Data production and results The uMngeni catchment in South Africa is under severe stress to supply the

drinking-water demand to two large residential areas, Durban and Pietermaritzburg, and several smaller ones. The four large reservoirs in the catchment are at risk of becoming eutrophic, mainly due to nutrient emissions

from agricultural activities, aging sanitation infrastructure and scattered informal settlements that lack basic sanitation infrastructure. These fast growing informal

settlements together with inadequate infrastructure in some areas contaminate the streams with raw sewage. The situation is expected to aggravate due to climate change, which will result in temperature and evaporation increases in the

catchment, and subsequently worsen the water quality.

To determine the potential changes in water quality under a future climate, a hydrological modelling approach was taken. A hydrological model was selected and then configured for the current conditions (Step 1). Future climate change

scenarios were then selected and climate files generated for the hydrological model (Step 2). Following this, the model was run using potential future climates

and an analysis undertaken (Step 3). Using these results a showcase which displays the results in an easily understandable manner for multiple stakeholders

was developed (Step 4). The last aspect undertaken was dissemination and communication (Step 5) to stakeholders.

The overall aim was to assist Umgeni Water, and at a more local scale the irrigation boards and conservancies in the catchment, to better understand how

the water availability and quality will change temporally and spatially through the uMngeni catchment under future climate scenarios.

4.1 Step 1: Data Collection and Model Configuration Description:

The greater uMngeni catchment, KwaZulu-Natal, South Africa, which is under severe stress to supply the drinking-water demand for two large residential areas (Durban and Pietermaritzburg), and several smaller ones was selected as a case

study for this project. The four large reservoirs in the catchment are at risk of becoming eutrophic, mainly due to nutrient emissions from agricultural activities,

aging sanitation infrastructure and scattered informal settlements that lack basic sanitation infrastructure. These fast growing informal settlements together with inadequate infrastructure in some cities contaminate the streams with raw

sewage. Public health and safety are affected when bacteria counts exceeded 400 per 100 ml (SA Guidelines); the actual concentrations can reach up to 1 000

000 counts per 100 ml. To investigate the potential changes in water quality of

STEP 1: Data collection and

model configuration

STEP 2: Defining

scenarios and CIIs

STEP 3: Scenario

modelling and analysis

STEP 4: Development of Showcase

STEP 5: Dissemination

and communication


C3S_422_Lot1_SMHI – D5.1.1B | 8

the uMngeni catchment under a future climate the ACRU Agrohydrological model

was selected. The ACRU model is a physical-conceptual, multi-purpose, multi-layered soil water budgeting, daily time step model (Schulze, 1995; Schulze et

al., 1997; Schulze and Pike, 2004; Warburton et al., 2012). The multi-purpose nature of this model allows for the simulations of many hydrological and terrestrial processes such as runoff, crop yield, reservoir yield, irrigation

demand/supply, evaporation and land use (Schulze, 1995, Schulze, 2000; Warburton et al., 2012). ACRU can be used as a lumped model, or as a

distributed model in large catchments with complex soils and land uses, whereby the catchment is disintegrated into small catchments of similar properties (Schulze, 1995). The ACRU model is sensitive to changes in land use and climate

(Schulze et al., 1995), and has been extensively used in South Africa (e.g. Warburton et al., 2010).

Using South African national sources of climate, land use and soils data (CWRR, UKZN Quinaries database), the existing configuration for the ACRU

agrohydrological model for the uMngeni catchment was updated. The ACRU model was run in distributed mode, with 1003 hydrological response units, for

each there was a climate file, land use, soils and streamflow response variables. For reporting purposes, those were clustered to 145 catchments and 13 Water

Management Units (WMUs). Baseline historical scenarios were run (1960 – 1999) and the simulated streamflows compared to observed flows.

Results: The conceptualization of the water budget in the ACRU model is shown in Figure

1. Precipitation, which is the driver of the hydrological cycle, is the major input component of the system. A certain percentage of this precipitation is initially abstracted as either stormflow or interception, and the remaining water is

infiltrated into the topsoil (A horizon), and once field capacity is reached it is further percolated into the subsoil (B horizon) as saturated drainage. If the

subsoil then becomes saturated, water continues to percolate further down the soil profile, into the intermediate zone and finally reaches the groundwater, contributing to runoff as baseflow. The total evaporation component includes the

evaporation of water from intercepted surfaces and from the soil and also includes the transpiration by plants, extracting soil water from the root zone.

Potential reference total evaporation, which can be estimated using the A-pan reference evaporation, is incorporated into the model to estimate the actual total evaporation. The generation of simulated runoff depends on the antecedent soil

moisture status and the rainfall intensity. The antecedent soil moisture deficit is thus, simulated at a daily time-step in order to assess the stormflow generated

following each individual rainfall event. A certain proportion of generated stormflow does not reach the outlet of the catchment on the same day as the rainfall event. Stormflow is, therefore, separated into various stormflow

components in the ACRU model, these include (a) quickflow, which is the proportion of stormflow that is generated and reaches the outlet on the same

day as the rainfall event; and (b) delayed flow, which is the proportion of stormflow that reaches the outlet, contributing to streamflow, several days after the rainfall event (Schulze, 1995). Ultimately, the ACRU model uses the

catchment information (e.g. land cover information), in combination with various methods and equations, to simulate the streamflow (and other output variables

such as sediment yield, crop yield, evaporation, peak discharge, rainfall and soil moisture) and hence, the hydrological response from a particular catchment.


C3S_422_Lot1_SMHI – D5.1.1B | 9

Figure 1: Representation of the water budget in the ACRU model (Schulze, 1995; Schulze

and Smithers, 2004)

For the purposes of this study the uMngeni catchment was delineated into 13 Water Management Units (WMUs) (Figure 2). These were further delineated into

145 subcatchments, with each being further subdivided into homogenous land use units. These homogenous land use units were routed to represent the flow of

water through the catchment (Figure 3). For each of the homogenous land use units the relative parameters to represent that land use were input (parameters from Schulze, 1995; land cover determined from National Land Cover 2000).

Further to that the climate information, soils data and streamflow response variables were input. The climate information comprised of daily maximum and

minimum temperature obtained from Schulze and Maharaj (2004), and daily rainfall obtained from Lynch (2004). The potential evaporation was calculated using the Hargreaves and Samani (1985) daily reference potential evaporation

equation. The soils data included the depth of the top and subsoil horizons, and the permanent wilting point, field capacity and porosity of the top and subsoil

horizons (obtained from the South African Climatological and Agrohydrology Atlas 2007).

With the model configuration and a model run under a historical climate complete, the next step was to define the future climate scenarios that would be

run.


C3S_422_Lot1_SMHI – D5.1.1B | 10

Figure 2: Water Management Units of the uMngeni catchment, in KwaZulu-Natal,

South Africa

Figure 3: An example of cascading (i.e. routing) of flows between subcatchments in

the uMngeni catchment and homogenous land use units within each

subcatchment


C3S_422_Lot1_SMHI – D5.1.1B | 11

4.2 Step 2: Defining scenarios and CIIs Description: With the ACRU agrohydrological model configured for the current conditions in the uMngeni catchment, the next step in the methodology was to define the

potential future climate scenarios and develop the required climate files for input into the ACRU model. The ACRU model requires, at the minimum, daily rainfall,

and daily maximum and minimum temperature. Further to this, the model was developed such that a “driver” climate station is selected to be representative of the climate of each subcatchment. The climate information needs to capture the

rainfall statistics at a point, rather than being a gridded average over a large area. Thus, the future climate files could not be gridded averages. Thus, the

decision was taken to employ the commonly used “change factor” method (Fowler et al., 2007; Chen et al., 2011; Forbes et al., 2011; Park et al., 2011) of statistical downscaling to derive the future climate files for input to the ACRU

model from the available GCM information.

Given that the uMngeni catchment is already under severe pressure, the decision was to model for the period 2011 – 2040. The CIIs for monthly maximum and minimum temperature and precipitation for the 19 available Global Climate

Models at 0.5 degree resolution for both the RCPs 4.5 and 8.5 scenarios for the 2011 – 2040 time period for the grid cells within the uMngeni catchment

produced within C3S_422_Lot1_SMHI contract were used. These CIIs were calculated with bias adjusted CMIP5 data available in the Climate Data Store (CDS) catalogue. For the rainfall, the CII precipitation (seasonality) which is the

mean monthly values of daily precipitation averaged for a 30 year period, expressed as a relative change was obtained from the C3S_422_Lot1_SMHI

contract for the period 2011 – 2040 for the available 19 GCMs and the ensemble mean for the grid cells covering the uMngeni catchment. The change factor was

then applied to the daily observed rainfall records of nine driver rainfall stations across the uMngeni catchment for the control period (1971 - 2000) to obtain the future climate scenarios. For the minimum and maximum temperatures, the CII

temperature maximum (seasonality) and temperature minimum (seasonality) which is monthly values of daily maximum or minimum temperature (at 2m

height) averaged over a 30 year period was obtained from the C3S_422_Lot1_SMHI contract for the period 2011 – 2040 for the available 19 GCMs and the ensemble mean for the grid cells covering the uMngeni catchment.

This index is given as absolute change against the reference period of 1971-2010 (future period minus reference period).

The CII future values were applied to the observed daily records to produce adjusted climate files for use in the ACRU agrohydrological model representative

of the projected climate for 2011 – 2040. Both the RCP 4.5 and RCP 8.5 scenarios were considered. The change factor method of downscaling is the most

widely used method for impact analysis (Chen et al., 2011), however, its limitation is that it assumes that the variability of the future climate will be same as the control observed climate (Forbes et al., 2011) and in applying the method

relative climate changes are assumed to be more important than absolute changes. This downscaling was necessary given the diversity in the uMngeni

catchment. For example, along the red line in Figure 4, a distance of 17 km in the Karkloof WMU, there is a 1000 m change in altitude, a 250 mm change in


C3S_422_Lot1_SMHI – D5.1.1B | 12

Mean Annual Precipitation (MAP) and a 5 degree change in Mean Annual

Temperature (MAT).

Figure 4: Example illustrating the range in altitude in the Karkloof WMU, uMngeni

catchment relative to the resolution of the GCM pixels

Results: The driver rainfall station for each of the subcatchments was assigned to one of

the six 0.5 degree grid cells covering the uMngeni catchment, coded M1 to M6. Following this, the extracted change factors for minimum temperature, maximum temperature and precipitation were applied to the observed historical (1970 –

1999) records for the subcatchments driver station to create a 36 future climate files for that subcatchment (18 GCMs; 2 emission scenarios).

The minimum temperature change factors ranged from an increase of 2.36°C to a decrease of 2.71°C, with the average change factor being 0.86°C. The

maximum temperature change factors ranged from an increase of 3.76°C to a decrease of 0.88°C, with the average change factor being 1.47°C illustrating that

in general the GCMs were indicating a larger increase in maximum temperatures than minimum temperatures. The percentage changes in precipitation ranged from 185% increase to a 63% decrease with the average change being 3%. The

large percentage changes can be misleading as these are during the dry winter months in the catchment where monthly precipitations are less than 5 mm on

average.


C3S_422_Lot1_SMHI – D5.1.1B | 13

4.3 Step 3: Scenario Modelling and Analysis Description: Given that the uMngeni catchment is already under severe pressure, the decision

was to model for the period 2011 – 2040. The climate files representative of 2011 – 2040 were used as input to the ACRU model and simulations were

undertaken for each of 19 GCMs for both the RCP 4.5 and RCP 8.5 scenarios. The outputs from the 2011 – 2040 simulation were compared to the simulated outputs from the historical baseline run. The modelled output compared included

temperature, precipitation and streamflow at a monthly scale, both the median and range of percentiles were compared.

Results: Only the RCP 8.5 scenario results were included in the Showcase, thus only those

graphs are included here. The maximum temperatures for the catchment will increase by 1 - 3ºC by 2040 with the ensemble mean indicating an increase of 1

– 1.5ºC by 2040 for both RCP4.5 and RCP8.5 (Figure 4). Although all models clearly show an increase in the maximum temperatures, there are fair differences between the various GCMs. This is a similar message that is given for maximum

temperatures at the global scale.

The increase in minimum temperatures for the catchment is less than those projected for maximum temperatures. The increases in temperature are up to 2ºC by 2040 under a RCP8.5 scenario (Figure 4) and 1.5ºC by 2040 under a

RCP4.5 scenario. These increases are less than those indicated at the global scale. Furthermore, these increases are less than those shown in the regional

downscaled projections produced by the South African CSIR (e.g. Archer et al., 2018).

At a global scale, it is indicated that precipitation will clearly decrease across southern Africa. For the grid cells within which the uMngeni catchment falls, the

message is not as clear. Under the RCP4.5 scenario, 6 GCMs indicate no substantial change in rainfall, 5 indicate a decrease of up to 10% by 2040 and 8

an increase of up to 10% by 2040. Under the RCP 8.5 scenario, 10 GCMs indicate an increase of up to 10% by 2040, 6 a decrease of up to 10% and 3 no substantial change (Figure 5). It is clear that there is still uncertainty

surrounding the likely rainfall changes for the catchment. Based on the ensemble mean and range between the 25th and 75th percentile by 2040 there does appear

to be a slight decrease in the summer rainfall months of January and February, with slight increases in November and December (Figure 5).

The changes in the accumulated streamflows simulated using the future climate files (2011 – 2040) show an amplification of the changes in precipitation due to

the non-linearity of the hydrological cycle. Decreases in streamflow are evident in the months of February, March and April for all the WMUs (Figure 6). The decreases in streamflow are due to the slight decreases in precipitation and

warmer temperatures which enhance evaporation. However, the range between the 25th and 75th percentiles is fairly large, indicating the uncertainty in the

change in accumulated streamflow which is expected as precipitation is the primary driver of the hydrological cycle and uncertainties still exist in the projections of precipitation for the future.


C3S_422_Lot1_SMHI – D5.1.1B | 14


C3S_422_Lot1_SMHI – D5.1.1B | 15


C3S_422_Lot1_SMHI – D5.1.1B | 16

Figure 4: Changes in air temperature for the thirteen WMUs in the uMngeni

catchment by 2040


C3S_422_Lot1_SMHI – D5.1.1B | 17


C3S_422_Lot1_SMHI – D5.1.1B | 18

Figure 5: Changes in precipitation for the thirteen WMUs in the uMngeni catchment

by 2040


C3S_422_Lot1_SMHI – D5.1.1B | 19


C3S_422_Lot1_SMHI – D5.1.1B | 20

Figure 5: Changes in accumulated streamflows for the thirteen WMUs in the uMngeni

catchment by 2040


C3S_422_Lot1_SMHI – D5.1.1B | 21

What is clear is that the temperatures in the catchment will increase in the

future. This will have impacts on the water quality of the catchment through an increase in water temperature and an increase in evaporation. An increase in

water temperature implies a broader spread of water borne diseases and pathogens. A water temperature change may also affect the multiplication rates of the pathogens concerned. Not only will the spread be wider, into areas which

cannot currently sustain the pathogens, but also the rate at which the reproduction will occur in the areas currently infected will increase, thus creating

more severe infestations. The increase in spread, reproduction rate and longevity of the pathogens will increase the risk of human disease on an already vulnerable set of communities living close to or reliant on the water courses. Increased

evaporation implies a decrease in the water available to dilute pollutants, thereby increasing the concentrations.

4.4 Step 4: Development of Showcase Description: Following the analysis of the results, the showcase was developed. The showcase

focused at the uMngeni WMU scale for ease of understanding. The historical and future (2011 – 2040, RCP 8.5) median monthly minimum and maximum temperatures (Figure 4), precipitation (Figure 5) and streamflow (Figure 6)

graphs for each of the WMUs were included, with the 25th to 75th percentile ranges shown for the future climate. The potential impacts of these changes on

water quality were interpreted and conveyed using an infographic. Results:

The showcase was developed for a range of stakeholders from lay people associated with the local scale conservancies in the catchment to high level

scientist at Umgeni Water. The showcase started with an introduction to the uMngeni catchment, including the climate and the current pressures on water

quantity and quality in the catchment. This leads onto a tab where the WMUs in the uMngeni were displayed on a Google Earth image. By clicking on a WMU a pop-up screen opened on which the changes in air temperature, precipitation and

accumulated streamflow could be viewed (example given in Figure 7 for the Lions River WMU). This allows for the changes to be viewed spatially through the

catchment and different WMUs to be compared to determine hot spot areas of change (if they had existed in the results).

The last aspect of the showcase was an infographic explaining the potential impacts of the projected changes in air temperature, precipitation and

streamflow on water quality (Figure 8).


C3S_422_Lot1_SMHI – D5.1.1B | 22

Figure 7: Example of the screen on the showcase where impacts of climate change are shown


C3S_422_Lot1_SMHI – D5.1.1B | 23

Figure 8: Water quality infographic produced for the uMngeni catchment


C3S_422_Lot1_SMHI – D5.1.1B | 24

4.5 Step 5: Dissemination and Communication Description:

The dissemination of the findings to the stakeholders has been through presentations at local meetings; distribution of links to the showcase to the

relevant stakeholders; and findings and results to the relevant stakeholders. Results:

The uMngeni catchment has, due to population growth, migration, land use shifts and financial and political pressures very pressing needs in terms of water and

sanitation provision for the here and now, and so many of the infrastructural development plans in place are to address changes proceeding at a far faster rate than that of Climate Change. Further to this, having recently struggled

through a nationwide severe drought, short term management has been the focus of Umgeni Water recently. Given these pressures, the philosophy of

Umgeni Water, therefore, is to continue developing infrastructure based on the historical and projected demand information but to “keep an eye” on climate change so that mitigation measures can be included as soon as it becomes

apparent that the information is well informed from an improved GCM science and more certainty at the catchment and local scale.

The stakeholders, including Umgeni Water, within the uMngeni catchment have identified several possible adaptation measures to be considered. The adaptation

strategies relevant to the Umgeni Water and potential changes in water quality under a future climate include preserving the upper areas of the catchment area

in a pristine state (or as close to as possible), relooking the flood line demarcation (particularly the 1 in 100 year flood), redesign of infrastructure, review of water quality discharge standards, reserve determination, assess solid

waste management, incorporate climate change into policy and decision making.

5- Conclusion of full technical report The case study used the Climate Impact Indicators for minimum and maximum

temperature and precipitation for the 19 available GCMs for both the RCP4.5 and RCP8.5 emissions scenario for the period 2011 – 2040 from the Copernicus

Climate Change Service. Using these CIIs the projected changes in climate by 2040 for the uMngeni were assessed and used to simulate projected changes in streamflow. These results were presented visually at a relevant spatial scale to

assist Umgeni Water and local level stakeholders in the catchment to understand the potential impacts of climate change on the water quantity and quality of the

uMngeni catchment, and the spatial distribution of those impacts, by 2040. From the results, it is clear that the air temperatures for the catchment are going

to increase by 2040 with the maximum temperatures rising more than the minimums. There remains uncertainty in the changes in the precipitation, both in

the amount and direction of change across the GCMs and this filters through to the streamflow as precipitation is the primary diver of the hydrological cycle. However, it does appear likely to be a decrease in streamflow in the late summer

months of February, March and April. There is no clear spatial pattern in the projected changes over the catchment, with the changes for most WMUs being

similar.


C3S_422_Lot1_SMHI – D5.1.1B | 25

References

Archer, E., Engelbrecht, F., Hänsler, A., Landman, W., Tadross, M. & Helmschrot, J.

(2018) Seasonal prediction and regional climate projections for southern Africa. In:

Climate change and adaptive land management in southern Africa – assessments,

changes, challenges, and solutions (ed. by Revermann, R., Krewenka, K.M.,

Schmiedel, U., Olwoch, J.M., Helmschrot, J. & Jürgens, N.), pp. 14–21, Biodiversity &

Ecology, 6, Klaus Hess Publishers, Göttingen & Windhoek. (14) Seasonal prediction

and regional climate projections for southern Africa.

Chen, J, Brissette, FP and Leconte, R. 2011. Uncertainty of downscaling method in

quantifying the impact of climate change on hydrology. Journal of Hydrology 401(3-

4):190-202.

Forbes, KA, Kienzle, SW, Coburn, CA, Byrne, JM and Rasmussen, J. 2011. Simulating the

hydrological response to predicted climate change on a watershed in southern Alberta,

Canada. Climatic Change 105(3-4):555-576.

Fowler, HJ, Blenkinsop, S and Tebaldi, C. 2007. Linking climate change modelling to

impacts studies: recent advances in downscaling techniques for hydrological

modelling. International Journal of Climatology 27(12):1547-1578.

Schulze, RE. 1995. Hydrology and Agrohydrology: A Text to Accompany the ACRU 3.00

Agrohydrological Modelling System, Water Research Commission, Pretoria, South

Africa.

Warburton, ML, Schulze, RE and Jewitt, GPW. 2010. Confirmation of ACRU model results

for applications in land use and climate change studies. Hydrology and Earth System

Sciences 14(2010):2399-2414.

D422Lot1.SMHI.5.1.1B: Detailed workflows of each case-study on … · 2019-02-18 · the streams with raw sewage. The situation is expected to aggravate due to climate change, which

Documents