INDIAN CROP YIELDS: HALF OF THOSE IN THE DEVELOPED WORLD
This video was created within our team to highlight the problems we saw on the ground with respect to poor soil health in rural India. With 50% of all Indians employed in agriculture, this is a huge issue to tackle; and one we're excited to take on.
NEED FINDING AND CHALLENGING ASSUMPTIONS
The first thing I wanted to do was ask questions..."Is soil testing even a need?"... "Perhaps there is a more pressing need?"..."Who are the key players?"...After six months of research and a trip to India, I came to the conclusion that there is a huge need for an affordable, usable, available, and actionable soil fertility diagnostic; one that could increase crop yields by empowering rural Indian farmers with the information about their soil health as well as practices they would need to take to improve it.
GOVERNMENT EXTENSION SERVICES: INEFFICIENT AND CORRUPT
None of this project would be relevant if the agricultural extension services of the Indian government performed as expected. Unfortunately, though, as with many government agencies in India, these extension services (called KVK's from their Hindhi acronym) are underfunded, understaffed, and innefficient. Thus, their reach is minimal.
FIRST ORDER OF BUSINESS: MEET THE TARGET CONSUMER
Early on, we identified a few promising NGOs in the agriculture space and established a good rapport with them. Then, we asked them to show us around and meet a diverse set of rural indian farmers - from women to men, rich to poor, etc etc. We conducted interviews with over 70 farmers gauging their attitudes and experiences with regards to soil testing, as well as what problems they face on their farms.
Group interviews were conducted by sex: men and women. We found that separating the sexes provided better data due to women feeling more open to share information when with their own sex.
We also conducted individual farmer interviews, which proved highly insightful as well.
A compendium of our farmer interview photos.
SO NOW THAT WE'VE TALKED, HOW DO WE ENGAGE YOU?
Interviews were great. But we realized we needed something more. So, with the immense help of our NGO partners, we came up with the idea of a design workshop - aimed at having farmers perform specific activities to answer critical questions. A video of our approach can be found here.
The best way to make use of the farmers' time was to invite a large group of them to partake in a series of design workshops. The idea was to expand on the interviews and have the farmers engage in design activities centered around critical questions. For instance, for the question of "what product attributes to the farmers most prefer" we created a pictographic card exercise highlighting attributes such as accuracy, longevity, or cost - and farmers then organized their card decks to establish a hierarchy of product attributes important to them.
Product attributes translated into Kannada for the farmer's reference.
Image of the different colorimetric (and distance) sensor variations we gave to the farmers. Farmers had a set of these sensors and had to fill out a sheet with the amount of fertilizer they had to add for each chemical given the colorimetric result. In the end farmers found the colorimetric strips rather confusing.
Women groups interpreting our paper protoytpes.
SENSORS AS COW FOOD?!
We also made small looks like props and prototypes to gauge how farmers interacted with different technology types. Here, a passive sensor implantable in the ground - which farmers found unattractive because of the high number they would need, and more surprisingly, because they were afraid their cows would it them!
A series of photos from our 8 design workshops across North and South India.
DESIGN THINKING APPLIES TO ACADEMIC RESEARCH
Research and design are analogous, starting with searching for inspiration from divergent fields. While butterflies (SEM below) have nothing to do with soil health, the fact that the irridesent color of their wings results from their microstructure, inspired me to think of nano-imprinted surfaces for soil diagnostics. I also considered: (1) colorimetric dye based sensing (3) plasmonic sensing (4) electrochemical sensing (5) and even synthetic bacteria sensing! In the end I decided electrochemical approaches were most appropriate for the user needs.
WHY SCREEN PRINT WHEN YOU COULD INKJET PRINT?
As I was screen printing my electrodes, I uniformity and efficiency became an issue. I tried to think of a new manufacturing method for my electrode elements. Alas, I found an optimized silver nanoparticle solution that could be inkjet printed through a thermal inkjet printer, much like the common HP desktop printers. And walla! Hundreds of sensors, each identical to the other, with just the click of a button.
A closeup under the microscope of the inkjet printed nanoparticles that serves as electrodes for the device. Science is beautiful!
The potentiometric three electrode setup of the sensors.
After inkjet printing a carbon nanotube transducer layer and dropcasting the ion-selective membranes, the devices are left to condition in different salt solutions overnight.
INKJET PRINTING THE WHOLE DEVICE CIRCUITRY?
Realizing the power of inkjet printing, we realized we could also inkjet print the whole device - both the disposable sensors and the voltmeter which measured the voltages from each chemical channel. Here is shown the inkjet printed "PCB" of the reader on paper.
"EXPERIENCES" LIKE MODEL
With the fully inkjet printed model, complete, we made a video to teach farmers how to use our "experiences-like" sensor prior to the workshop. Here, we tried to gauge the user experience as accurately as possible with this prototype: (1) taking a composite soil sample (2) adding a soil extractant solution (3) stirring and waiting (4) dipping in the ion selective electrodes (5) waiting for a response (aka lights to go on, highlighting a numeric code that represents the soil health data) and (6) combining the numeric code with the alphabetical code of the crop species from the back of the sensor (together, the alphanumeric code would be texted out to a central service provider).
With this model, we got a lot of incredible feedback from the farmers; highlighting small things like we never would have been able to discover at home - like the brightness of the lights being too much for indoor spaces in india, or the preferred use of different lights for different chemicals.