How to make an iPoster with usual paper media

Most researchers aim to present their work at conferences, often considering an oral presentation as the ideal outcome. Conversely, being assigned to present a poster can sometimes be seen as a second choice. However, I believe this perception overlooks the unique opportunities posters provide. Interacting with attendees at a poster session often leads to more in-depth conversations than those following an oral presentation.

Recognizing this potential, some conferences, such as the European Geosciences Union (EGU), have innovated with new formats like the PICO (Presenting Interactive COntent) sessions. These sessions enhance the traditional poster experience by allowing researchers to present their work on touch screens, making the interaction more dynamic and engaging than paper posters. This approach has gained popularity and expanded over time.

Unfortunately, smaller conferences may not have the resources to provide touch screens for such interactive experiences. As an alternative, I've experimented with a method that partially recreates the PICO experience. The idea is to make the poster more visual and less text-heavy while incorporating QR codes that link to short videos explaining different parts of the poster. These videos can cover the abstract, references, and explanations of figures.

Implementing this iPoster strategy involves some technical steps: recording the videos, uploading supplemental material for immediate access, and creating QR codes. Recording videos is now straightforward; you can use a virtual meeting platform to record yourself presenting the poster. For hosting the videos and additional materials, I prefer using the Open Science Framework (OSF), which allows for the visualization of PDFs, videos, and Jupyter notebooks without needing to download them first. Videos can also be uploaded to platforms like YouTube or Vimeo.

Once you have your materials uploaded, you can generate QR codes using one of the many available online tools. Embed these QR codes into your poster at relevant points. This way, visitors can access the supplementary material directly from their mobile devices while viewing your poster, even if you are not present.

This approach enhances the traditional poster session, making it more interactive and informative. While there is always room for improvement, this method can significantly enrich the experience for both presenters and attendees.

Clicking on the Figure above you can see and download the iPoster Example. This was actually derived from this poster. For a more generic/general post on how to do poster, please give a look to this blog post. 

Everything is improvable !

GEOframe-New AGE material for beginners

Dear User or Dear Explorer,

Here we aim to summarize some of the material related to GEOframe-NewAGE. The main source is certainly the GEOframe blog:

• The GEOframe blog 

However, for a logical introduction, it may be useful to start here:

• GEOframe Essentials 

The most recent material on GEOframe-NewAGE is from the latest school, accessible from this point:

• The 2024 GEOframe Winter School

By following the links for each day, you can download the slides and watch the lesson videos. 

• and previous Schools and material

Another useful resource is provided through hydrological modeling tutorials:

• Hydrological Modelling 2023

GEOframe's infrastructure is based on the Object Modelling System v3:

• OMS3 essentials

Various papers and applications related to GEOframe have been written and developed; you can find them here:

• GEOframe blogs posts in AboutHydrology

For any further assistance, the GEOframe crew can be reached at geoframe-schools@googlegroups.com.

Please feel free to reach out  us if you have any questions. Next Winter School  on GEOframe-NewAGE will be in January 2025 from 7 to11 in Trento University.  Next Summer School (on Land-Atmosphere interactions will be June 2-6 2025.  This Summer will be holding one Summer School on GEOframe-NewAGE in El Cairo and one in Mumbay (both the last weeks of July).

And finally we talk about isotope concentration

 If I understand correctly, colleagues arre measuring the concentration of an isotope in stream or some other catchment outlet.  This is quite not a measure of the age of water but the latter has to be inferred from  a hypothesis the system behavior, and possibly the knowledge of the distribution of water ages within the control volume, obtained iteratively from an initial data (derived from the isotopic data of precipitation).

Therefore, assume you have measured in the outflow a concentration of 5 of solute X in appropriate units. You mai have  the knowledge of the concentration of the same solute in the water of a given age resident in the control volume. It would be a tuple like (7,3,6,2) where the number of concentration is the number of precipitation happened before the present instant. The 5 units out of the tuple can be taken all from the oldest (7) concentration, from the youngest (3,2) where the 3 are out of the 6 of the penultimate concentration.

Therefore a concentration of 5 can be fitted with a water age of 4 in the first case or (3*2+2*1)/5=1.8 in the second case.

In fact, the concentration doesn't identify  the exact age of the water but rather a range of ages spanning from the "first in first out" age to the "last in first out" age illustrated in the simple example.

Potentially the old water is infinitely old and, therefore, the lower bound could be very large, while the upper bound for the ages is, clearly, related to the last rainfall age.

So, the division between young water and old water is reasonable. Instead of trying to guess the distribution, you're trying to guess parts of its integral.

Then SAS (StorAge Selection functions, see this previous post for understanding what they are) are summarizing the mechanics of mixing inside the control volume. If the waters mix a lot the travel time and the residence time should coincides  because of the uniform selection of isotopes (or tracers) among the various populations. SAS have a non trivial form if the mixing of waters is incomplete.

Due to the localized nature of water physics, the actual travel times are intricately linked to the specific location from which water emerges or is extracted from the soil. Within the unsaturated zone, it's logical to posit that newly introduced water is the first to flow out for two primary reasons: it occupies the empty pores, and these pores are typically where water is loosely retained. However, when young water mixes with old water in the saturated zone, the aquifer's pressure rises, causing a pressure wave that pushes out the oldest water near the streams.

Consequently, in a typical zeroth-order catchment, runoff tends to adhere to a "last in, first out" (LIFO) logic, while subsurface stormwater tends to expel old water first, following a "first in, first out" (FIFO) logic. Consequently, the recession of the flood wave is expected to contain a higher proportion of old water than young water. In this hypothetical simplified model delineating two distinct dynamics, and drawing from the aforementioned insights, it becomes feasible to estimate the global effect of the system, as proposed by Rigon and Bancheri in 2022 and applying the recent methods of TT probability determination.

Vegetation withdrawal we can think to extract water from the places where active roots are, according to root density. Plants with shallow roots should mostly sample young water, while plants with deeper roots wold sample older water but maybe also being effective in mixing water of different ages because of exudation and because of preferential water infiltration close to the roots.

The presentation (under construction) is exemplifying these concepts.

Large Language models in Earth Observation

If you've grasped the concepts discussed in the previous post, you're likely poised to explore how to effectively integrate them into your hydrological endeavors. However, you're not starting from scratch. Institutions like ESA and NASA have already outlined some of their applications through a series of posts and contributions. Machine Learning undoubtedly plays a significant role in Earth Observation and remote sensing analyses (see for instance here for an overview) or browse the activities of the RSLab for a perspective from the University of Trento. 

Here they are a list of interesting links that I collected:

• NASA and IBM Openly Release Geospatial AI Foundation Model for NASA Earth Observation Data
• Integrating Earth Observation (EO) with Large Language Models (LLMs): Towards A Multimodal EO-Language Model
• Linking Large Language Models for Space Domain Awareness (The topic can look like quite different from our central one but the connection is the management of large amount of data that should be merged for decision making)
• Large language model agent to interact with Earth Observation Data 
• Imagining the Future of Large Language Models and Open Science

After having read them I found that those links are give suggestions but at the end kind of work around the technical questions which I am looking for. 

I am sure that there exist many other resources on this topic, so, please do not hesitate to point my attention to them and I'll make the list grow.