Recent searches
Search options
Oh LOOK at all the solar panels on roofs in this industrial area in Pakistan!
(It's not the Barents Sea, Google Maps just switches longitude and latitude for some reason).
Anyway the answer to the question 'why do you think there is lots of solar in Pakistan when official stats say there isn't" is a. customs data and b. I can see it from space.
Btw Pakistan's power grid is 46 gigawatts, serving 236 million people. The US has 333 million people and a power grid of about 1,300 gigawatts.
Pakistan imported about 13 gigawatts of solar modules in the first half of 2024.
I worked with Atlas (https://atlas.co/) to use their SolarEye machine learning program to find solar rooftops in Pakistan. It found 443 projects occupying 27 million square metres, but visibly missed many more than it found.
What it does confirm is that the projects are clustered in industrial areas, and are therefore probably on the roofs of factories with little involvement with the grid.
Pakistan's energy regulator, NEPRA, notes that power consumption is down 9.1% year on year in 2023. NEPRA attributes the drop mainly to high power prices cutting economic activity and making residential consumers curb consumption, with rooftop solar only a third factor.
But NEPRA doesn't know how much solar the country has, either. We think it has about 12.7GW of solar already (compared with 50GW on-grid power capacity) and will add 10-15GW of solar in 2024.
This makes Pakistan the world's sixth-largest solar market in 2024, which is just wild.
(This is meant to be paywalled but the map is being spread around X, it's just that nobody on Mastodon can afford us).
Also, working with Atlas was great, best machine learning firm I have worked with to detect solar panels (and I've worked with a few, which was disappointing).
The results weren't *precise* (I think it missed 2-3x as many as it caught, and detected nothing small) but it's a problem where some answer is better than none, and at least Atlas was efficient and their interface was really good. This is, alas, not a problem worth $$$$ to anyone.
@solar_chase Would the IEA not be open to at least partial funding of such a study, under ‘infrastructure market growth monitoring’ or whatever label unlocks some money?
@BashStKid I guess that's up to the IEA.
There's also the legitimate question of "does knowing about it help". My clients are mainly interested in this from a perspective of global solar module demand, how that impacts the supply: demand balance for solar modules, and what it might do for prices (not much, there are still oodles of solar modules for the foreseeable future).