Remote sensing break out on introduction to sensing technology

session

#1

Remote sensors
Documentor didn’t know much about terms so please clarify where needed.

What people want and why they came to the group?
Assimilated climate data sets.
Types of remote sensing discussed?
Open access optical remote sensing
Light r processing tools.
Better idea how to use it?
What is people’s experience?
Laming ground based (sp) and satellite sensing.
Access remote sensing imagery climate data.
Earth engine a platform for planetary scale analysis motis, 200 m resolution, many climate data sets. Interested in using this type of data for land managers to understand carbon and water cycles.
Interested in knowing what is going on out there in terms of data sources.
Need a better understanding of how images are translated into estimates.
Terrestrial remote sensing getting a better sense in how work can help farmers.
New and emerging tech like drones and how those fit into products that are useful to smaller farmers. Integrating into research applied.
Learn more about satellite images are used in research and development.
Ground based sensors from remote locations wants to do uav work but how can I use it?
Data processing side. Sensing does not seem open source. So glad to see this chnging.
GIS experience how to make topographic maps evaluating water interactions on landscape
Applied geosolutions mostly here to talk about open tools that he has made and if people have developed tools to analyze remote sensing data.
What is the broader remote sensing community up to?

What do we do now?
Anyone interested in giving and introduction to remote sensing?
Someone has slides to share about remote sensing these will be shared on the forum by Sabrina.
Remote sensing is indirectly sensing systems.
People want to know what the work chain is.

Q and A for different types of sensor developers (The idea was for one developer to speak on their technology then we would ask questions. This did not quite work.)
Bio-geo-solutions: Processing and remote sensing for publicly available data sets
Go in and get access to the api and download.
Get access to GIPS then a driver, standard set of functions for getting the data
Make a driver that took the pre CIP data from a data set and you can put in what if questions.
Common thing in remote sensing is you have an area of interest and you have a question you pick a resolution and it makes a picture you have to align it once it is aligned you can look at time series or specific parameters.
Parameters such as Vegetation “ndvi”
Make a model based on that parameter
For analysts to use the data, apply the models, not as far as earth engine scale
Question:
When integrating multi resolution how do you choose the resolution?
Depends on the if you are interpolating if you interpolate and go to highest resolution rather than starting at the lowest it is an estimate rather than direct measurement.
The direct measurement is used as a picture or point in space comes from the amount of light scattering into a lense at a point in time.
The location of that information is more of an average.
The resolution choice is start course and go fine.
The 30 M resolution is encouraging conservation by coming up with a system to verify if people are using conservation practices. Encourage the types of practices you would like to see
How often do you get a pass/ how often is data collected?
5 days
Remember if it is overcast you might go several months without data. A strike out was more common in the past because there are more satellites now.
There are also some workarounds for clouds. Using a thermal band to look through clouds.
Are there satellite data not affected by cloud cover?
Yes radar, they can see through clouds but don’t have the same qualities.
It can measure distance in a 3d space. Can sense different materials.
Material volume, what is reflecting.
If you have clear optical and clear radar they can complement each other.

What are different uses for this technology?
Radar is good for crop or no crop but it will not be able to differentiate between crop type.
Everything is contrast. The other bands come into play with Planet imagery- estimating yields of certain areas.
Verification by seed companies if people are sticking to their contractual agreements.
Some companies would audit what producers are doing in late 90s-2000s.
Remote sensing could be for external industries only? NO! but it is descriptive not diagnostic. It is up to you to understand what that image does. You will get information but it is up to you how to use it.
Being used for prescriptive problems in the fields in variable rate technology (precision ag). Cheaper rate of crop insurance if using this data to influence inputs.
Using layers to add information.
Everything is not always quantitative analysis.
Farmers might use it qualitatively by looking at wet spots and where tile and drainage will go on new or old land.
Smaller farms it may not be as useful. Might not need to use it but might want to use it to map areas.
Optical was hit hard in this discussion! But you can also get elevation data from 30 M freely available.
Reflectance data to identify glyphosate resistant palmer amaranth from crop research is currenly in development. But it’s not there yet.
Light R processing tool is convenient stacking. Published in Git hub
All of these ideas might not work everywhere. Everything gets tested in the amazon, which is easy compared to more subtle vegetation.