Machine Learning To Save Meghalaya’s Forests
Discover how Machine Learning through data, algorithms, and conservation efforts are coming together to save Elaeocarpus prunifolius a rare species from extinction in Meghalaya.
Well, here’s a tale that’ll knock your wellies off — a bunch of boffins in India have turned to machine learning (yes, the same tech that’s recommending your next binge-watch) to save a rather shy and retiring resident of Meghalaya’s forests: Elaeocarpus prunifolius. Now, before you mistake it for an exotic curry or a lesser-known Harry Potter spell, let me assure you — it’s a tree. And not just any tree, but one that’s tottering on the edge of extinction, like a tipsy uncle at a wedding.
Published in the ever-so-serious journal Modeling Earth Systems and Environment, the study details how these scientists, armed with nothing more than algorithms, ecological niche modelling, and presumably a strong brew, have managed to map out nearly 9% of Meghalaya as potentially suitable habitat for this endangered chap of a tree. Talk about finding a needle in a haystack — or in this case, a root in a rainforest.
“Machine learning allows us to forecast suitable habitats with remarkable accuracy,” said co-author Dibyendu Adhikari, presumably while sipping tea and staring knowingly at a spreadsheet. “Traditional methods were a bit all over the place — rugged terrain, not enough field data, you know the drill.” Indeed, the hills of Meghalaya aren’t exactly a stroll through Hyde Park.
Their model — clever little thing — pinpointed around 201,321 hectares where E. prunifolius could potentially thrive, with a special nod to about 10,361 hectares in the East Khasi Hills. That’s not just good news — that’s the botanical equivalent of finding a pub open past eleven in a sleepy village.
But — and there’s always a “but” — life on the ground is no walk in the arboretum. Illegal logging, urban sprawl, and agriculture are muscling in faster than you can say “chainsaw massacre”. Even in protected areas, the trees seem to be giving up faster than a commuter in a Monday morning queue. Seeds are dropping early, rodents are having a nibble, and insects have turned the place into a buffet. It’s not so much survival of the fittest as survival of the luckiest.
Between 2014 and 2018, a rather grim field survey found just 275 adult trees across 43 sites. Throw in 828 saplings and 1,252 seedlings, and it begins to sound less like a thriving population and more like a very small reunion of distant cousins.
But hang about — it’s not all doom and gloom. The machine learning models did spot a glimmer of hope. Turns out, not all small forests are hopeless cases. In fact, the seedlings seem to rather like undisturbed patches, suggesting these leafy enclaves might just be nature’s last strongholds — like the last good chippy in a gentrified high street.
The researchers — Viheno Iralu, Krishna Upadhaya, and Hiranjit Choudhury — aren’t just throwing data into the wind. They’ve proposed a proper action plan: seed banks, tree nurseries, and strict anti-deforestation laws (hear that, loggers?). Places with delightful names like Lum Shynna, Lawbah Arliang, and Krang Suri forest have made the shortlist for restoration. One hopes they fare better than most committee-approved plans which, let’s be honest, often end up gathering more dust than momentum.
So there we are — science, nature, and a smidge of digital wizardry teaming up for a cause. Elaeocarpus prunifolius might have been on the ropes, but with a little help from the geeks and a lot less from the loggers, this tree might just leaf its troubles behind.
Cheeky puns aside — here’s hoping.
What's Your Reaction?
