Post navigation

Procedural textures: A perfect match for environmental factors

Three cross sections through a terrain using procedural textures for factor visualization in 3D-space

Three cross sections through a terrain using procedural textures for factor visualization in 3D-space

Understanding how spatiotemporal patterns of environmental factors relate to phenomena we observe in nature is a fundamental quest in environmental systems sciences. The complexity of these patterns makes visualization of factor gradients and dynamics particularly useful.

Fortunately, in Blender and in other 3D content creation software we can make use of so-called procedural textures, i.e. material definitions for objects that are continuously calculated through 3D-space using mathematical algorithms. This way it is possible, for example, to visualize patchy nutrient availability throughout a block section of a terrain. For this a specific noise texture can be parameterized. A simpler linear gradient texture can be used for water tables and the like.

To visualize more complex factor interactions overlays of different procedural textures and different blend modes for the colors can be used. Ultimately, for very specific or flexible pattern visualization Blender provides material components that can use material definitions of the powerful Open Shading Language.

Will EVOs materialize in future?

A 3-D printed model „landscape“ for illustrating the metastability concept. Depending on the location of release, kinetic energy and disturbances the spheres come to rest at troughs of different potential energy or perhaps even on a saddle (labile position).

A 3-D printed model „landscape“ for illustrating the metastability concept. Depending on the location of release, kinetic energy and disturbances the spheres come to rest at troughs of different potential energy or perhaps even on a saddle (labile position).

2013 is the year 3D printing pentrates everyone’s mind. New 3D printers and fields of application pop up every week. Thus, the question arises whether EVOs will remain virtual for teaching or wether some of them will turn into physical objects eventually.

An answer to this question should consider two main criteria: In which cases do physical rather than virtual EVOs have a didactical advantage and provide easier access to understanding a concept, and are the efforts for 3D printing an EVO justified (if at all possible)?

It’s probably too early to tell, but the options are on the table and the options will likely proliferate in future …

Tech preview: Investigation EVOs on Android devices

An Investigation EVO prototype running on a 7'' Android tablet

An Investigation EVO prototype running on a 7” Android tablet

As most students use smartphones or tablets these days, the question comes up whether “EVO-apps” should be developed for such devices. As smartphones are almost always at hand, such Apps would give students great freedom when to work with EVOs.

Well, at least for the more demanding Modeling EVOs such devices are too limited in many respects. However, Investigation EVOs with their game-like approach may be particularly well suited for being used on tablets or smartphones if they provide on-screen user interfaces and touch manipulation. A player app for Android has been developed that allows running Blender game-engine based EVOs on Android devices. The app still lacks a few important features like text rendering but otherwise is already mostly functional.

We consider including the Android platform for future development of Investigation EVOs.

The first link in a chain: computer graphics research

The EVO project is all about using the open 3D content creation and animation software Blender for developing Environmental Visualization Objects. But for many sophisticated visualization functions in Blender it’s not the Blender developers who made the start. Rather, these functions originate in more generic computer graphics research that tries to find useful algorithms for visualization and simulation problems in the three spatial dimensions plus time.

Smoke plume rendered in Blender

A good example is the smoke simulator that was implemented by Blender developers Daniel Genrich and Miika Hamalainen based on a research paper by computer graphics scientists of Cornell University and ETH Zurich. And there you have it, a chain of links: basic computer graphics research linked to implementation in open software linked to use for specialized application development (e.g. a cloud development EVO) linked finally to students exploring weather dynamics at the end of the chain.

So, what does the future have up it’s sleeve for EVO creation? An intriguing place for some possible answers is the Interactive Geometry Lab led by Prof. Olga Sorkine. Check it out …

The first feedbacks are in.

Students have worked with the very first Modeling EVO! Using the R statistics package and the Experimental Design EVO they performed virtual growth experiments and related their results from statistical analysis to observations they made in the visualized experiment.

The feedbacks by the students were consistently positive: They liked the flexible approach to “run” an experiment with instantaneous visual outcome and to be able to switch back and forth between statistical results and inspecting the experimental system. Click on the image to have a closer look at some of the comments …

Innovedum event 2012 at ETH Zurich

Innovative projects and approaches for teaching at ETH within the innovedum framework were presented in the hall of the Environmental Systems Sciences building on 7 Nov. The EVO project took part in the exhibition featuring the scope of the project (poster and showreel) and demonstrating the new Modeling EVOs.

 

Lively discussions at the EVO booth

Quite a number of people stopped by and many of them were intrigued by the potential the EVO approach offers. We had fun discussing further options for future EVOs and making connections to colleagues from other departments during the exhibition and the dinner afterwards.

 

EVO presentation at the Blender Conference 2012

The EVO project was presented at the 2012 Blender Conference in Amsterdam (12 to 14 October). We explained the motivation behind our EVO project and its general concept and demonstrated case examples on how to use Blender for creating Modeling EVOs. The presentation inspired a lively discussion, and we could make first contacts with colleagues at other academic institutes. Furthermore, we took the chance to discuss implementation of possible further simulation features in Blender directly with the developers.

A pdf of the presentation (3.5 MB) can be accessed here.

Welcome to our EVO project!

We develop EVOs (Environmental Visualization Objects) as teaching tools in environmental systems science.

To let colleagues and interested people follow this project we have just started this blog and a whole site dedicated to providing infos and resources about everything related to EVOs. Well, right now it’s just a start. But we hope this EVO site will grow and reflect the unfolding project as well as the perception of the (academic) community active in environmental sciences: What is the role EVOs could play in communicating environmental science in the widest sense?

evapotranspiration

Evapotranspiration EVO: Linking functional relationships to observed changes in the system