Tygron for Scientific Research and Subsidence calculation: Difference between pages

From Tygron Support wiki
(Difference between pages)
Jump to navigation Jump to search
No edit summary
 
 
Line 1: Line 1:
=Use-case 4: ''“the Virtual River Game”'' =
The subsidence model calculates the gradual downward settling of the ground's surface where a peat layer is present. The considered subsidence takes place due to oxidation and compaction of the peat layer.
''' ''This page of the manual is still under development. The final version will be launched on April 8th 2021 during the [https://www.tygron.com/en/event/edu-event-2021/ EDU event 2021].'' '''
The total subsidence is currently calculated as the result of oxidation and compaction over the years. Oxidation is dependent on the ground water level and the clay thickness. Compaction is dependent on the toplayer thickness, the peat fraction and raised surface terrain.  


A PhD candidate at the University of Twente was assigned to create a serious gaming environment for river interventions as a tool that facilitates stakeholder engagement. The research was one among 20 projects as part of the [https://kbase.ncr-web.org/rivercare/projects-overview/ RiverCare research program]. The researcher is a human-centered interface designer, and is not an expert in river interventions. After developing and testing several prototypes, the researcher created the the virtual river game, a unique board-game that is digitally connected to the Tygron platform. To simulate river interventions, players can move the pieces on the board, and immediately see the changes on the Tygron screen including the new map, as well as numerical calculations and graphs.
The ground water level is initialized by either raster or surface area data and optionally adjusted each year (indexation). When indexation is active, the surface water level is adjusted for the settling each year and the ground water levels will change accordingly.


[[File:virtualriver3.png|50px|frame|center|The board component of the Virtual River Game]]
The calculation model is configured by the subsidence overlay wizard and it's result are visualized with the [[Subsidence (Overlay)|subsidence overlay]].


==Content of the Research==
==Calculations==
During a calculation step, the following aspects are calculated in order:
* The temperature at the start of the year is calculated
* Based on that, [[#Yearly recalculation of Parameter A|parameter a]] of the oxidation formula is calculated
* The [[#Oxidation calculation|oxidation subsidence]] is calculated
* The [[#Compaction calculation|compaction subsidence]] is calculated
* The surface water levels are lowered by the amount of subsidence times the [[#Indexation|indexation]]
* The [[#Ground water depth|ground water level]] is lowered based on subsidence and surface water level indexation
* The new ground water level serves as input for the next calculation step


===General information===
===Yearly recalculation of Parameter A===
{{:Parameter a recalculation formula (Subsidence Overlay)}}


*'''The PhD Position:'''"[https://kbase.ncr-web.org/rivercare/projects/project-g1/ A serious gaming environment for river interventions]"
===Ground water depth===
{{:Ground water depth formula (Subsidence Overlay)}}


*'''Main Researcher:'''"[https://people.utwente.nl/r.j.denhaan?tab=contact Robert-Jan den Haan]"
===Oxidation calculation===
{{:Oxidation formula (Subsidence Overlay)}}


*'''Research Programme''': [https://kbase.ncr-web.org/rivercare/projects-overview/ RiverCare]
===Compaction calculation===
{{:Compaction formula (Subsidence Overlay)}}


*'''Department''': [https://www.utwente.nl/en/et/dpm/ Department of Design, Production & Management]
===Subsidence calculation===
{{:Subsidence formula (Subsidence Overlay)}}


*'''Educational Institute''': [https://www.utwente.nl/en/et/ Faculty of Engineering Technology] of the [https://www.utwente.nl/ University of Twente]
===Indexation===
{{:Indexation formula (Subsidence Overlay)}}


*'''Involved Organisations''': [https://www.rijkswaterstaat.nl/ Rijkswaterstaat], [https://www.deltares.nl/en/ Deltares], [https://www.witteveenbos.com/ Witteveen+Bos], [http://www.arcadis.nl/ Arcadis], [https://www.hkv.nl/en/ HKV], [https://www.txchange.nl/ T-Xchange], [https://www.nwo.nl/toegepaste-en-technische-wetenschappen-ttw TWO] & [https://www.tygron.com/nl Tygron Geodesign Platform]
===Terrain height===
The height of the terrain can be manipulated during a session by stakeholders taking a land sculpting [[action]], creating [[open water]], or constructing [[levee]]s. These actions will result in settlement, which can be found under the Settlement result type. Creating or demolishing [[construction]]s generally do not change the height of the land, and do not result in changes in the settlement results.


===Background of the researcher===
==Configuring overlays==
The researcher did not have experience in GIS or geodesign or coding
The subsidence overlay has a number of ways to configure it. Both values which serve as input for the overlays directly, as references to attributes of areas which provide input for the calculations.
Limited programming/coding base + course along the way.
Design methods
Practical side - making the board - CAD laser cutting.
Masters in Industrial Design Engineering (utwente)
===Objectives and desired outcome===


The main objective of the research is to create a virtual game that allows players to interact with an imaginary river section.
===Keys===
The subsidence overlay has a "Keys" tab in the [[right panel]] in the editor. Most keys are [[attribute]]s of [[area]]s. When the overlay calculates, it will look per [[Grid overlay|grid cell]] for the existence of these attributes.


The game would allow different stakeholder groups to:
{{Overlay keys|suppresscategory=true|
*learn about each others' perspectives regarding river management
{{:Water level (Subsidence Overlay)}}
*take better decisions of river interventions
{{:Water level output (Subsidence Overlay)}}
*reflect upon the potential effects of those interventions
{{:Indexation (Subsidence Overlay)}}
{{:Subsidence (Subsidence Overlay)}}
{{:Clay thickness (Subsidence Overlay)}}
{{:Peat fraction (Subsidence Overlay)}}
{{:Toplayer thickness (Subsidence Overlay)}}
{{:Drainage (Subsidence Overlay)}}
}}


To do that, the researcher is expected to:
Besides these attributes, 2 more model parameters can be configured.
*develop and test a serious virtual game where players experience how river systems function.
*write a dissertation paper


{{anchor|Years}}
'''Years'''<br>
The amount of years to simulate during the calculation, in 1-year steps. It is possible to set this value anywhere between 1 to 1000. This parameter can be configured with the [[Years (Subsidence Overlay)|YEARS]] model attribute of the Subsidence overlay.


==Research process==
'''Ground Water Tiff'''<br>
If the option to use a Ground Water Tiff is checked, a GeoTiff can be selected to use for the ground water levels. You can either use any of the provided default GeoTiffs, or upload and use your own. <br>


The game was developed based on a need-driven approach
Three ground water GeoTiffs covering the Netherlands are available by default. These GeoTiffs are a combination between a high resolution GeoTiff containing ground water levels for rural areas and a low resultion GeoTiff containing ground water levels for city areas, where data from the low resolution GeoTiff was only used to fill the gaps in the high resolution GeoTiff.
Who was involved:
One of the available ground water GeoTiffs, relevant for the Subsidence, is the Mean Lowest Watertable (MLW, or GLG in Dutch).
The supervisors were involved - professor in human-centered design + professor in Hydraulic engineering
post-doc involved for discussions during the process
As for the game: Florian + Hansje for the feedback
Practically, connecting the board to the game: Rudolf
Deltares: Fedor Baart (co-author) - coding expertise and modeling expertise
Other researchers


===Attributes===
The subsidence overlay also has [[Model attributes (Subsidence Overlay)|model attribute]]s. All attributes have a default value, but can be changed to configure the subsidence calculation.


===Time distribution===
{{overlay attributes|suppresscategory=true|allowselflinks=true|
 
{{:Parameter a (Subsidence Overlay)}}
Eng of 2014 - End of 2019
{{:B (Subsidence Overlay)}}
 
{{:C (Subsidence Overlay)}}
The research was designed for a 4 year programme, however the researcher did it within 5 years, on a 4-days a week basis, since he was employed by the university 1 day/week.
{{:Climate start temp (Subsidence Overlay)}}
In this case, the initial research was very time-intensive, taking up to 3 years. This phase included intermediate prototypes.
{{:Climate final temp (Subsidence Overlay)}}
3 prototypes were formally tested.
{{:Climate soil temp factor (Subsidence Overlay)}}
The rest (2 years) were the development of the game (prototyping) and testing.
{{:Climate oxidation (Subsidence Overlay)}}
The testing included test sessions with stakeholders. 5 sessions were organized at the end of the research phase, and all took place within 6 weeks.
{{:Default clay thickness (Subsidence Overlay)}}
Afterwards more sessions were organized but unfortunately they stopped due to the covid lockdown.
{{:Default peat fraction (Subsidence Overlay)}}
 
{{:Default top layer thickness (Subsidence Overlay)}}
===Resources & guidance===
{{:Hi passive drainage (Subsidence Overlay)}}
 
{{:Low passive drainage (Subsidence Overlay)}}
====Tygron====
{{:Years (Subsidence Overlay)}}
The researcher also relied on the web-based form of the API as well as the wiki.
}}
Tygron was involved from the beginning of the research.
Tygron was the most applicable tools for this game.
(T-xchange has a toolkit for a game design, but Tygron seemed more suitable)
 
The researcher had to learn how to use Tygron
He received some training from the Tygron team.
He visited the Tygron office for 2 full days in the beginning.
Part of it was a discussion about the project: What can and cannot be done with Tygron.
That was followed by a few visits to discuss the links between Tygron and the board game.
 
Data sources in the game: Delft3D Flexible Mesh hydrodynamic model, the BIOSAFE biodiversity model and a self-developed cost model based on prices from research.
 
[[File:VirtualGame2.jpg|thumb|250px|right|The Virtual River Game's hardware and software.]]
 
[[REST_API| API (Application Programmers Interface)]]
[[File:VirtualGame1.jpg|frame|center|The physical set-up of the virtual river game including the physical board, touchscreen monitor, projector, and webcam.]]
 
====Board Game====
 
Coding - a lot of coding with Python as well as GIS
Researcher relied on others for help with coding.
Interviews with stakeholders in river management.
Used solidworks to develop the physical board and the icons.
Used the university laptop to create it, but dedicated the mini computer that could be carried.
During that time, Tygron was developing the water module, so the researcher instead used a water model (Delft3D model) from Deltares which was CPU intensive.
For Tygron, the laptop was okay, but the water model from Deltares was heavier
A student assistant was hired to help with the creation of the board
Colleagues and fellow researchers were also involved in the design and the evaluation phase (testing)
Other 7 were involved in the research.
 
 
Budget was dedicated for mostly the dedicated computer, materials for the board, touch screen monitor, projector, webcam, wood material, etc.. as well as hiring the student assistant.
 
==Outcome of the research==
 
. As a concept idea we started to make a physical game board that represents a fictional and abstract part of the Dutch river. The game board is a spatial area divided into equal hexagonal locations. Each location is filled by two modular types of game stones: height and land use. For example, there are low stones to form the main channel of the river. Slightly higher stones, together with different land use, form stones as agricultural use, natural grassland and forest the floodplains. Even higher stones form the dikes.
 
We have developed a game table and software around the game board that makes the physical game board digital. Through additional operations, we make the digital board suitable for the Delft3D Flexible Mesh hydrodynamic model, the BIOSAFE biodiversity model and a self-developed cost model. The digital board is also used to control the Tygron Platform. For example, players see the game board as a virtual world on the platform. Where the land use is agricultural on the board, players see cows grazing in the virtual world. Where the land use is forest on the board, the players see trees.
 
The outcome was already envisioned from the beginning- the game is not only the game but also where you play it. (digital + analogue)
It was not necessarily the vision of the research consortium. They had a more scientific background and their vision was focused on getting a very realistic and complicated game, as close to reality as possible, however the researcher's vision was to make something based on the need, something that is useful.
 
The initial prototypes, the researcher translated real location into a board(east of nijmegen)  
For the final game, they took the characteristic of a dutch river and used it as a representation of a typical dutch river. (width, slope, depth, dykes'heights, etc.) So it is not a real location.
 
[[File:Video01introduction.png|500px|link=www.youtube.com/watch?v=KzwTWI7hKag|video1]]
[[File:Video02Visualization.png|500px|link=https://www.youtube.com/watch?v=7JtL2o3sHHQ|video2]]
 
 
You can find the full research paper here: [https://www.sciencedirect.com/science/article/pii/S1364815220309129 The Virtual River Game: Gaming using models to collaboratively explore river management complexity]
 
You can find the summary poster of the project here: [[File:Summary poster - virtual river game.pdf]]
 
==Feedback and recommendations==
-The researcher recommends 3rd parties to create added value (what is actually needed)
- Also involving 3rd parties is important for practical expertise and knowledge
 
Difficulties: the connection was a technical challenge - trial and error
Expectation of the project partner of a very complicated looking game was a barrier
Develop prototypes early on, even if it is only 2/10, then you can discuss on a baseline instead of something conceptual
Once the prototype was playable, a baseline was created and discussions started.
 
Expectation management.
Spend more time on thinking what should be developed, what could be done, what should not be done, what is possible, etc.
 
Some of the things were new for Tygron especially the connection
 
==Contact==
For more information about this PhD research, you can contact us at: info@Tygron.com
 
You can also contact [https://people.utwente.nl/r.j.denhaan?tab=contact Robert-Jan den Haan], researcher at the University of Twente at: r.j.denhaan@utwente.nl
 
{|class="wikitable" style="margin: auto; background-color:#ffffcc;"
| [[Tygron for Spatial Planning|<<Previous page]]
| [[Tygron for Decision-Making|Next page>>]]
|}

Revision as of 10:47, 15 March 2021

The subsidence model calculates the gradual downward settling of the ground's surface where a peat layer is present. The considered subsidence takes place due to oxidation and compaction of the peat layer. The total subsidence is currently calculated as the result of oxidation and compaction over the years. Oxidation is dependent on the ground water level and the clay thickness. Compaction is dependent on the toplayer thickness, the peat fraction and raised surface terrain.

The ground water level is initialized by either raster or surface area data and optionally adjusted each year (indexation). When indexation is active, the surface water level is adjusted for the settling each year and the ground water levels will change accordingly.

The calculation model is configured by the subsidence overlay wizard and it's result are visualized with the subsidence overlay.

Calculations

During a calculation step, the following aspects are calculated in order:

  • The temperature at the start of the year is calculated
  • Based on that, parameter a of the oxidation formula is calculated
  • The oxidation subsidence is calculated
  • The compaction subsidence is calculated
  • The surface water levels are lowered by the amount of subsidence times the indexation
  • The ground water level is lowered based on subsidence and surface water level indexation
  • The new ground water level serves as input for the next calculation step

Yearly recalculation of Parameter A

The a parameter is recalculated yearly, to account for a changing climate. The new value for the a parameter is obtained by multiplying the original Parameter a is with a factor. This factor is calculated based on an interpolated average temperature for that year, a soil temperature factor, micro activity and an oxidation factor.

The value of the a parameter is updated each year as follows:

Delta temperature is calculated as:

Delta micro activity is calculated as:

Delta subsidence is calculated as:

The new parameter a value for year y is calculated as:

where

= The a parameter for the current year.
= The a parameter value at the start of the calculation, defined by the model attribute A.
= The current year. The first year is defined by START_YEAR.
= The start year of the simulation, defined by CLIMATE_START_YEAR.
= The final year of the simulation, defined by the addition of the start year and YEARS.
= The average temperature of the start year, defined by CLIMATE_START_TEMP.
= The average temperature of the final year, defined by CLIMATE_FINAL_TEMP.
= The Q10 factor, configured as 3. This value is currently not adjustable.
= Soil temperature factor, defined by CLIMATE_SOIL_TEMP_FACTOR.
= Oxidation factor, defined by CLIMATE_OXIDATION.

Ground water depth

At the start of a simulation, the ground water depth is initialized with the ground water depth GeoTIFF (if provided) and is optionally overwritten by (managed) water areas' water level.

Furthermore, the ground water level can be managed with drainages (provided as underground buildings), either actively or passively.

Additionally the terrain height can change due to subsidence that occurred in previous years and due to actions taken that raised the terrain. Managed water areas can react to these changes when indexation is configured. For indexation, see indexation formula.

The following formulas describe how the yearly adjusted ground water depths are obtained.

Ground water level managed by drainages

where

is the calculated ground water depth of a grid cell at year y.
is the ground water depth actively maintained by a drainage
is the ground water depth of the managed water area.
is the area water level adjustment for year y.
is the ground water level passively maintained by drainages.
is the ground water level increase calculated based on the are water level adjustment due to indexation and the subsidence in the previous year.

Ground water level managed by water areas

The effect of changes of managed surface water levels (by water areas) on the ground water level. The x-axis indicates the relative depth of the managed surface water level. The y-axis indicates the meters of change to the ground water level, per meter change in surface water level. The smaller the relative depth is, the less the ground water level equalizes with changes in the managed surface water level.

The ground water level can also be managed by water areas, which control the surface water levels. To make sure ground water levels do not rise (too much), surface water levels in the water areas can be lowered automatically with respect to the subsidence. This process is called indexation. The new ground water level can then be estimated based on the indexation on the managed water level of surface water and the calculated subsidence of the previous year. The increment in ground water level is rarely exactly the adjustment of the managed surface water level and is often less, depending on the relative depth of the surface water level. The following formulas have been developed to estimate the increase of the ground water level.

The new relative surface water level depth is calculated as:

Next, based on the old and new depths, the ground water depth change is divided into three sections, each which its own rate of contribution.

Finally, the ground water depth change is calculated and the new ground water depth is obtained:

where

is the current relative water depth of water area a.
is the adjusted relative water depth of water area a.
is the water level adjustment for indexation of current year y.
is the subsidence that occurred up to year y.
is the calculated adjusted ground water depth for year y, based on the occurred subsidence and indexation of previous years.
is the initial ground water depth at the start of the simulation.
is the estimated change of the ground water level based on occurred subsidence and indexation.

Oxidation calculation

The amount of subsidence due to oxidation is calculated by the following formula:

where

is the subsidence due to oxidation.
is the calculated ground water depth for year y in cell c.
is the calculated a parameter value for the current year .
is the clay thickness of the terrain in cell c, defined by CLAY_THICKNESS
is the b parameter value, defined by the model attribute B
is the c parameter value, defined by the model attribute C

This formula was provided by experts, who have derived this formula empirically.

Compaction calculation

The amount of subsidence due to compaction is calculated by the following formula:

where

is the calculated subsidence due to compaction.
is the peat fraction of the respective grid cell c.
is the top layer thickness of the respective grid cell c.
The terrain height increase as a result of the actions taken during a session, such as the creation of levees.
is the number of days in a year times the current year being calculated.
is the first regression constant (0.015853041) for the effect of the peat fraction and time in days.
is the first regression constant (0.006617643) for the top layer thickness and time in days.
is a regression constant (0.200468677) for the change in surface height, for example caused by added materials.
is the second regression constant (0.02348519) for the effect of the peat fraction.
is the second regression constant (-0.010061616) for the top layer thickness.

This formula is based on provided expert data in the form of a reference table, indicating the amount of subsidence based on the parameters used in the formula above. The formula's results conform to the reference table to within an average of a tenth of the margin of error of the original table.

Subsidence calculation

The total subsidence for year y is calculated as:

where

is the peat fraction of the respective grid cell c.
is the top layer thickness of the respective grid cell c.
is the maximum possible subsidence in year y.
is the total subsidence in year y
is the subsidence due to oxidation in year y.
is the subsidence due to compactions in year y.

Indexation

Indexation is the policy of managing the surface water level such that it keeps the nearby ground water levels (more or less) at the same depth.

A water area which is fully indexed (1.0 = 100%) will have its surface water level lowered by the same amount as the terrain height has lowered due to subsidence. Since it lowers just as much as the terrain itself, the ground water level(s) relative to the surface of the land will remain the same.

In a water area which is not indexed (0%) the surface water level remains at the same level. Any subsidence taking place will lower the land, and thus reduce the relative water depth. An area indexed by 50% will have the surface water level lower by half of the amount of subsidence.

To make matters a bit more complex, the change in surface water level rarely changes the ground water level with the same amount. Instead, the amount the ground water level changes is estimated using the ground water depth formula for water areas.

The indexation formula is as followed:

where:

is the average subsidence for water area a.
is the subsidence in grid cell c for year y.
is the number of non-water grid cells within water area a.
is the amount the water area is changed in meters.
is the indexation fraction for area a.

Terrain height

The height of the terrain can be manipulated during a session by stakeholders taking a land sculpting action, creating open water, or constructing levees. These actions will result in settlement, which can be found under the Settlement result type. Creating or demolishing constructions generally do not change the height of the land, and do not result in changes in the settlement results.

Configuring overlays

The subsidence overlay has a number of ways to configure it. Both values which serve as input for the overlays directly, as references to attributes of areas which provide input for the calculations.

Keys

The subsidence overlay has a "Keys" tab in the right panel in the editor. Most keys are attributes of areas. When the overlay calculates, it will look per grid cell for the existence of these attributes.

Besides these attributes, 2 more model parameters can be configured.

Years
The amount of years to simulate during the calculation, in 1-year steps. It is possible to set this value anywhere between 1 to 1000. This parameter can be configured with the YEARS model attribute of the Subsidence overlay.

Ground Water Tiff
If the option to use a Ground Water Tiff is checked, a GeoTiff can be selected to use for the ground water levels. You can either use any of the provided default GeoTiffs, or upload and use your own.

Three ground water GeoTiffs covering the Netherlands are available by default. These GeoTiffs are a combination between a high resolution GeoTiff containing ground water levels for rural areas and a low resultion GeoTiff containing ground water levels for city areas, where data from the low resolution GeoTiff was only used to fill the gaps in the high resolution GeoTiff. One of the available ground water GeoTiffs, relevant for the Subsidence, is the Mean Lowest Watertable (MLW, or GLG in Dutch).

Attributes

The subsidence overlay also has model attributes. All attributes have a default value, but can be changed to configure the subsidence calculation.