-
Problem 1. Models for honeybee arrival and blossom phenology
Industry: Climate change
Industry Representative: Jennifer Fitchett, School of Geography, Archaeology and Environmental Studies, University of the Witwatersrand
Moderator:
Student Moderator:
Problem statement:
Phenology refers to the timing of annually recurrent biological events and their biotic and abiotic drivers. These phonological events are triggered by the change in seasons, as temperatures change and rainfall begins or terminates. Under climate change, the timing of these events is shifting. In many instances, events such as blossoming that marked the beginning of spring are now occurring in late winter. Each species, however, responds to a discrete selection of abiotic forces. For one species, the trigger may be the occurrence of daytime temperatures warmer than 20 degrees Celsius; for another it may be the timing of first rainfall. This results in a progressive mismatch between pollinators and predators and prey, as one species will arrive/hatch/emerge from hibernation long before its food supply is available. In this problem we are considering the plight of honeybees under climate change, as a result of the progressive advance of blossoming in South Africa. Models for honeybee arrival and blossom phenology will need to be developed and compared to calculate the threat of mismatch.
Supporting Material:
Bartomeus et al. 2011 (Study Group Problem 1 2020)
Burkle et al. 2013 (Study Group Problem 1 2020)
Chuine et al. 1999 (Study Group Problem 1 2020)
Frund et al. 2013 (Study Group Problem 1 2020)
Jacaranda Phenology SSAG (Study Group Problem 1 2020)
Kudo & Ida 2013 (Study Group Problem 1 2020)
Leong et al. 2015 (Study Group Problem 1 2020)
Petanidou et al. 2014 (Study Group Problem 1 2020)
Renner & Zohner 2018 (Study Group Problem 1 2020)
Singer & Parmesan 2010 (Study Group Problem 1 2020)
Wilmer 2012 (Study Group Problem 1 2020)
Presentation
Report-back presentation
-
Problem 2. Modelling the turbulent flow in Lake Kivu
Industry: Energy
Industry Representative: Denis Ndanguza, University of Rwanda, Rwanda
Moderator:
Student Moderator:
Problem Statement:
Turbulence is an irregular motion characterized by chaotic changes in pressure and flow velocity and which in general makes its appearance in fluids, gaseous or liquid. Generally, this is an irregular condition of flow in which the various quantities show a random variation with time and space coordinates, so that statistically distinct average values can be discerned. It is in contrast to a laminar flow, which occurs when a fluid flows in parallel layers, with no disruption between layers. Turbulence is encountered in most flows in nature and industrial application. Natural turbulent can be found in oceans, rivers, lakes and in the atmosphere, whereas industrial turbulent flows can be found in heat exchangers, chemical reactions, etc. Turbulence arises due to instability occurring at high Reynolds numbers. Turbulence modelling is essential in environmental ?ows, which comprise ?ows in rivers, estuaries, coastal seas and lakes. Previous researchers have shown that the Reynold number in Lake Kivu is high and this is a sign of turbulence existence. Down to a certain depth, turbulence is caused by waves and currents generated by winds and eddies caused by surface cooling. In Lake Kivu, this mechanism happens between 60 and 70 m of the lake.
Modelling turbulence in Lake Kivu is therefore of essential importance to the simulation of ?ow, the temperature (turbulent movement can spread the temperature) and biological activity in lake. Sometimes, the wide range of scales and apparently random nature of turbulent eddies make turbulence di?cult to model and a wide range of turbulence modelling approaches can be developed. Based on this motivation the issue addressed here is to apply any technique in fluid dynamics to model the turbulence movement in Lake Kivu which is a complex in term of stratification and stability.
Supporting Materials:
2020: Modelling turbulent flow in Lake Kivu - Surface Layer Hydraulics
Presentation
Modeling the turbulent flow in Lake Kivu
Report-back presentation
-
Problem 3. Juice holdup detection in a sugar cane diffuser
Industry: Sugar Cane Processing
Industrial Representative: Richard Loubser, Sugar Milling Research Institute NPC, c/o University of KwaZulu-Natal, Durban.
Moderator:
Student Moderator:
Problem Statement
The process for extraction of sugar from sugar cane in most South African mills involves a counter current washing process in a diffuser.
The best results are achieved with maximum wetting or juice holdup in the cane bed. The diffuser has sight glasses along the length of the diffuser for the operators to judge how much juice is held up in the bed. The operators then make adjustments to diffuser settings to change juice circulation patterns to compensate for changes in juice quantity that they see in the sight glass.
The sight glass is in a recess so there is a gap between the fibre of the bed and the glass surface. What the operator sees is a level of juice between the bed and glass. If this is a true level in the bed, it implies that the juice backs up from the diffuser screen to a height in the bed. Since juice flows down from the top of the bed and the diffuser screen at the lower boundary is perforated, the presence of a distinct liquid level is disputed. The level in the sight glass is often observed to drain away suddenly and then re-establish itself a short time later. The liquid level has not been observed in laboratory experiments using a cylindrical column or a rectangular tank with a static cane bed. This suggests that the level observed in full scale could be a boundary effect at the wall of the full-scale diffuser.
Since work is being done, using image analysis, to measure the position of the sight glass liquid level, it is important to know what this level represents. A model which relates the actual hold-up of juice in the bed; possibly including aspects such as the cane permeability, its variation, diffuser bed speed, diffuser bed height, juice feed to the top of the bed and juice distribution history, is needed to assess the relationship between juice hold-up, or cane bed saturation, and the juice level observed in the sight glass. This will help in the interpretation of the measured level in terms of the degree of holdup in the adjacent cane bed.
Presentation
202008 Juice holdup detection in a sugar cane diffuser (35MB)
Report-back presentation
Juice holdup detection in a sugar cane diffuser
-
Problem 4. Image analysis of sugar cane preparation
Industry: Sugar Cane Processing
Industrial Representative: Richard Loubser, Sugar Milling Research Institute NPC, c/o University of KwaZulu-Natal, Durban.
Moderator:
Student Moderator:
Problem Statement
The process for extraction of sugar from sugar cane in most South African mills involves a counter current washing process in a diffuser.
The first step is to break the cane stalk into fine pieces to expose the sugar containing juice for extraction. This preparation of the cane is done using a hammer mill shredder. After preparation, the shredded cane is fed to a diffuser where it is extracted using a counter-current washing process. Best results are achieved with maximum contact between the percolating juice and cane. Too much juice, however, leads to flooding and uncontrolled mixing of the juice with an associated loss in extraction.
Percolation performance of the shredded cane in the diffuser depends on the degree of preparation. If the cane is underprepared, it is difficult to wash the sugar from the cane. Overprepared cane will form a more densely packed bed in the diffuser with low permeability. The permeability influences juice flow patterns in the diffuser and low permeability can cause flooding of the diffuser.
The degree of preparation depends on the variety and growing conditions of the cane. The shredder is set to accommodate the average cane that is prepared. This leads to variation in the degree of preparation and hence the permeability of the cane bed. Currently cane preparation is measured using an off-line process. It would be useful if the cane preparation could be measured in real time allowing continuous adjustments to be made to the shredder clearances to compensate for variability in the cane.
A good front-end engineer would look at the prepared cane and be able to judge the degree of preparation. Can machine vision be used to achieve the same result?
Data was collected during an experiment at the beginning of 2019. Samples of cane with differing levels of brown leaf were prepared and then tested to give parameters which were designed to express percolation and extraction characteristics of the cane. The cane samples were also photographed.
Four cane varieties were used. Subsamples of each were subjected to the following treatments:
- Leaf was burned off the cane stalk
- Leaf was stripped by hand
- Medium amount of leaf was left on cane
- High amount of leaf was left on cane
Each treatment was analysed in triplicate as follows:
- Percolation rate: A given mass of cane was loaded into a column and percolation rate (hence permeability) was measured.
- Density: A given mass of cane was loaded into a tube and a fixed force applied to the cane with a plunger. After a fixed time, the volume was determined giving the density of the cane. (Density is strongly correlated to flow rate.)
- Displacement rate index: Cane was washed and the time constant for the mass transfer determined. Conductivity was used as an indicator of the mass transfer.
- Photograph: Three different photographs were taken of each batch of cane under similar lighting conditions.
The number of samples was therefore:
(4 varieties) X (4 treatments) X (3 replicates) = 48 (48 photographs).
This is a preliminary study is to determine if there is sufficient information in the photographs to determine whether it is feasible to use machine vision to extract the degree of preparation of the cane. If this shows a reasonable chance of success, techniques of taking photographs on a conveyor can be developed and more data collected for training a system.
Presentation
202008 Image analysis of sugar cane preparation
Report-back Presentation
Image analysis of sugar cane preparation
-
Problem 5. Green roofs to mitigate the Urban Heat Island
Industry: Climate Change
Industry Representative: Anne Fitchett, Assistant Dean, Faculty of Engineering and the Built Environment, University of the Witwatersrand
Moderator:
Student Moderator:
Problem statement
The Urban Heat Island is a phenomenon where the temperature of a region of a city (generally the inner city) is higher than the surroundings. This is caused by the prevalence of dense and dark coloured materials. The dense materials, such as concrete, absorb solar radiation in the daytime and store it until the evening, when the heat is released into the atmosphere by convection. Very dark materials, such as asphalt, absorb much more heat than lighter colours that reflect most of the solar radiation back into the atmosphere. The combination of these two types of materials cause the area of an inner city to be several degrees warmer than the surrounding areas, causing an up-swell of air that induces thunderstorms over the city, leaving the surroundings in a rain shadow.
There is a body of knowledge that suggests that the vegetation on green roofs can mitigate the Urban Heat Island through shading, insulation of the soil layer and evapotranspiration. Johannesburg inner city is built largely of reinforced concrete with plat roofs that are ideal for the installation of green roofs as no additional structural strengthening is required.
The problem is to model the existing Urban Heat Island in comparison with the natural terrain that would have characterised the pre-development state (taken as the temperature data from the Johannesburg Botanical Gardens). Various percentages of the area of the city with green roofs can then be modelled to determine an optimal area under green roof in the city.
Supporting Material
Urban Heat Island Numerical Model (Study Group Problem 5 2020)
Urban Heat Island Causes Effects and Mitigation (Study Group Problem 5 2020)
A diffusive Boussinesq plume (Study Group Problem 5 2020)
Presentation
Green roofs to mitigate the Urban Heat Island
Report-back presentation
Green roofs to mitigate the Urban Heat Island
-
Problem 6. Chessboard Waves
Industry: Shipping and Tourism
Industry Representative: Thama Duba, Durban, KwaZulu-Natal
Moderator:
Student Moderator:
Problem Statement:
In the popular tourist destination of the Isle of Rhe in France there are a set of waves that beautifully display the force of nature at its most powerful. The wave system showcases a strangely shaped square pattern looking like a chessboard that is perfectly displayed on the surface of the ocean. This is actually because the island is in an area where two seas meet-this is called a cross sea. Since the two seas that intersect in this area have different weather patterns and weather systems, the interaction of these waves forms squares.
The square waves caused by cross seas are very dangerous. This is because the rip currents and riptides are stronger than average in this phenomenon. The square waves can become a danger to swimmers and surfers and have caused boating accidents and ship wrecks.
The study group is asked to answer the following questions:
- What mechanism generates a square wave; for example, what type of waves would interact to form square waves?
- What mathematical model would best describe a chessboard wave? Is it a deep-water wave or a shallow water wave?
- What are the propagation properties of these waves?
- What makes these waves dangerous?
- Why is this phenomenon not being spotted at Cape Point?
References:
https://shareably.net/dangerous-square-waves
https:/en.wikipedia.org/wiki/Cross_sea
http://earth.eo.esa.int/cgi-bin/confsea10.pl?abstract=349
Videos of square waves: bing.com/videos
Presentation
202008 Mathematical model for chess board waves
