Ruishan retweeted
So beautiful, they melt my heart~
I tried my own take at street level with GPT-6.2 Astra: an Amsterdam canal corner and Shanghai's Anfu Road.
Built with real building data and seasons. ↓
april-zhu.com/pocketcity?cit…
Day 19. Senses: Smell
Smoke over Ukraine, Oct 2026: days a satellite saw heat in each 15 km hex. 7,871 detections, up to 24 of 31 days in one hex. Not confirmed strikes: field fires are included. #30DayMapChallenge
Data: NASA FIRMS, ERA5, SaveEcoBot
Ruishan retweeted
The U.S. bars Nvidia’s top Blackwell chips from export to China. So why do six Chinese models seem to bear the chips’ fingerprints? Using the models’ random starting numbers, a researcher connected them to the chips—and thinks this could expose even more.
sciam.com/article/chinese-ai…
At the #COP31 high-level segment, IUCN Director General Dr Grethel Aguilar called for a rapid, just and equitable transition away from fossil fuels, more finance, and faster protection and restoration of nature, with Indigenous Peoples, local communities, women and youth at the table.
bit.ly/4Cbt3Hb
#ClimateAction #IUCN
Ruishan retweeted
INFOGRAFÍA » @WSJGraphics analiza visualmente la imparable expansión global del poder chino: inconsolata.com/post/8327243…
Ruishan retweeted
Tropical sea surface temperatures (20N to 20S) are shattering the record for this time of year, at a full 0.5C warmer than the prior record (set in 2023). This is being driven by this year's record-setting El Nino event, on top of longer-term human-driven warming.
Ruishan retweeted
Building a digital forest using space based LiDAR.
Expanded this visualization to 50M+ GEDI lidar waveforms, revealing tree canopy heights across the entire state of Oregon.
Next up: the entire United States!
Exploring more space-based LiDAR data...
I mapped NASA’s GEDI lidar footprints (from 2019–2026) across 4 × 4 km near H.J. Andrews Experimental Forest, over USGS 1 m 3DEP terrain.
Then I used 2020 GEDI heights and Sentinel-2 imagery to model the canopy and render it as Gaussian splats.
Pretty fun result, but could be greatly improved with more stereo imagery.
Fun Fact: the tallest tree in Andrews Forest is 306 ft!
Ruishan retweeted
Climate tipping points can bring abrupt, potentially irreversible change. ESA's Climate Change Initiative uses satellite records and models to detect early warning signals of transitions such as Amazon forest dieback, dryland vegetation shifts and thawing permafrost.
Today at #COP31, the session ‘Prepared for Tipping? Adaptation and Decision Making Under Deep Uncertainty’ explores how to adapt under uncertainty.
🗓️ 1.15–2.45 p.m. UTC+3
📍 Side Event Room 4, AŞIK VEYSEL
📸© ESA/Planetary Visions
Ruishan retweeted
先日紹介したEmbeddingですが、期間を自由に指定できるカスタム版も近い将来使えそうです🛰️
これまでの年次データに加えて、季節ごとや月・週ごとの作成にも対応。自分のプロジェクトに合ったEmbeddingを作れるので、応用がますます増えそうです🙌
mapsplatform.google.com/reso…
Ruishan retweeted
"Türkiye İstatistik Haritası" yayında
Elimdeki tüm veri setlerinden coğrafi kırılıma sahip olanları tek haritada topladım. TÜİK, YSK, BDDK, MEB, YÖK, STB, ÇSGB, Diyanet gibi birçok kurumun verileri yer alıyor
600+ ölçümü inceleyin:
burakkocer.com/calismalar/in…
Ruishan retweeted
Satellite view in, 3D model out. It works for any place on Earth.
Get your model in 30 seconds
Ruishan retweeted
สรุปสั้นๆ เข้าใจง่ายแบบนี้ค่ะหายนะแรร์เอิร์ธคืออะไร ?
1. แรร์เอิร์ธคือแร่สำคัญ
เอาไปทำมือถือ รถไฟฟ้า ชิป ขีปนาวุธ แต่ตอนขุดต้องใช้สารเคมีแรงมาก มลพิษสูงมาก
2. จีนไม่อยากขุดในบ้านตัวเอง
เมื่อก่อนจีนขุดเอง 90% ของโลก พอธรรมชาติในประเทศพัง ก็สั่งปิดเหมืองในบ้าน แล้วหันมาลงทุนขุดในพม่าแทน ตั้งแต่ราวปี 2558
3. รัฐประหารพม่าทำให้เหมืองบูม
หลังปี 2564 พม่าวุ่นวาย กลุ่มชาติพันธุ์ในรัฐฉานคุมพื้นที่ได้ ต้องการเงินซื้ออาวุธ ทุนจีนต้องการแร่ เลยจับมือกันเปิดเหมืองหลายร้อยแห่ง
4. วิธีขุดคือทิ้งพิษไว้ให้คนอื่น
เจาะภูเขา อัดสารเคมีเข้าไปละลายแร่ แล้วส่งแร่ไปจีน 100% ส่วนน้ำเสียที่มีสารหนู ตะกั่ว แคดเมียม ปล่อยทิ้งไว้ตรงนั้น
5. ไทยรับพิษเต็มๆ
แม่น้ำสาย แม่น้ำรวก แม่น้ำกก ต้นน้ำอยู่รัฐฉาน สารพิษไหลเข้าเชียงราย เชียงใหม่ แล้วลงแม่น้ำโขงผลคือ น้ำขุ่น สีเพี้ยน ทำประปาไม่ได้ ปลาเป็นแผล คนไม่กล้ากินปลา ผักขายไม่ออก ท่องเที่ยวพัง
6. ทำไมแก้ยาก?
ต้นเหตุอยู่ในพื้นที่ที่รัฐบาลพม่าคุมไม่ถึง ไทยแก้คนเดียวไม่ได้ ส่วนจีนที่เป็นคนได้ประโยชน์ที่สุดยังนิ่ง ไม่รับผิดชอบ
สรุปประโยคเดียว:
จีนได้แร่ พม่าติดอาวุธได้เงิน คนไทยปลายน้ำรับสารพิษที่ตัวเองไม่ได้ก่อ
Ruishan retweeted
This is one of my favorite scientific figures in computational social science: the Division of Democrat and Republican Party members of the U.S. House of Representatives from 1949–2012. You can see our history in a figure resembling mitosis.
journals.plos.org/plosone/ar…
Ruishan retweeted
卧槽!中国两千多年的历史,被人做成了一张能拖动的动态地图!
GitHub 上这个开源项目叫《江山时序》,Star 已经拿到 1.9K!
战国一路到现代,18 个历史时期、164 张疆域快照全都覆盖在内。
最有意思的是拖一拖时间轴,秦灭六国、三国鼎立、唐宋元明清的疆域变化,都能亲眼看一遍。
古今地名、历史事件也能查,甚至能跟着李白、苏轼、玄奘的足迹在地图上走上一遭!
以前看历史全靠死记硬背,如今两千多年历史直接演给你看。
在线体验:deanchen.me/jiangshan-map/
GitHub:github.com/DeanChensj/jiangs…
Ruishan retweeted
.
Türkiye’nin enerji altyapı haritası
Ember tarafında hazırlanan çok güzel haritalar var. Alternatif bakış yararlı olabilir
Harita üç ana başlığı bir arada gösteriyor: Yenilenebilir enerji, petrol-doğalgaz altyapısı ve nükleer.
Öne çıkanlar
Hidroelektrik hala belkemiği. Atatürk, Keban, Karakaya, Birecik gibi büyük barajlar özellikle 2026’daki yüksek üretimle gaz ve kömür bağımlılığını azalttı.
Güneş rekor seviyelerde katkı sağlıyor. Karapınar başta olmak üzere güneş santralleri aylık elektrik ihtiyacının giderek artan kısmını karşılıyor.
Rüzgar daha çok batıdan geliyor (İzmir, Balıkesir, Çanakkale hattı).
Nükleer: Akkuyu-1 reaktörünün inşaatı tamamlandı, 2026 sonu itibarıyla devreye girmesi bekleniyor.
Petrol ve gaz tarafında ise Türkiye, hem transit ülke hem de kendi altyapısıyla kritik bir konumda. Türk Boğazları’ndan günde 3 milyon varilden fazla petrol geçiyor. TANAP, TürkAkım ve mevcut rafineri-gaz terminalleri bu ağı tamamlıyor.
Harita seçili önemli tesisleri gösteriyor, kapsamlı bir envanter değil. Ama Türkiye’nin enerji dönüşümünün ve jeopolitik konumunun özetini çok iyi veriyor.
Kaynak: S&P
A new initiative launched at #COP31 to speed up renewables, grids and storage while protecting nature and people.
AGENT brings together governments, companies, investors and Indigenous Peoples’ organisations to build energy infrastructure that minimises harm to biodiversity.
bit.ly/4iZTM25
Ruishan retweeted
Measurements show: we have increased the CO2 amount in the atmosphere by over half. The annual Global Carbon Budget report shows where the CO2 comes from and where it goes.
Don’t get fooled by false claims that isotopes show the increase is natural! 1/2
Ruishan retweeted
🇨🇳 China's high-tech manufacturing output surged 17.6% YoY in October. However, retail sales grew by just 0.4% and these two numbers perfectly capture one of the biggest paradoxes of the current Chinese economy.
📊 In October, value added in high-tech manufacturing increased by 17.6% YoY compared with just 4.8% for overall industrial production, a 12.8-percentage-point gap that has been widening since 2024. While China's overall industrial production has generally been growing at around 5/6%, its high-tech sectors have been accelerating. The sector-specific figures are equally impressive with +21.9% for integrated circuits, +26% for industrial robots and +46.8% in value added for manufacturers of specialized electronic equipment. I have to acknowledge that Beijing is successfully transforming its industrial base. China is expanding its manufacturing capabilities in important sectors and strengthening its global competitiveness.
📉 Meanwhile, in October, retail sales increased by just 0.4% YoY. Over the first ten months of the year, total fixed-asset investment contracted by 8.1%, real estate investment plunged by 21.6%, and private investment fell by 11.4%. China is expanding its productive capacity at an extraordinary pace while domestic demand remains extremely weak. While successfully modernizing its industrial sector, China is still struggling to rebalance growth toward household consumption. The prolonged property downturn continues to weigh on household wealth and confidence, while private businesses remain cautious about making new investments.
⚠️ How far can this model go? If China continues expanding its industrial capacity much faster than domestic demand, it will need to find an increasingly large share of its customers abroad. The thing is international markets are not unlimited and this strategy risks intensifying trade tensions with the United States and Europe. The more China strengthens its industrial power, the more dependent it may become on foreign demand if domestic consumption fails to keep pace, and that dependence becomes problematic in an environment characterized by rising trade barriers and aggressive industrial sovereignty policies.
China is demonstrating that it knows how to build the industrial base of tomorrow but can it build an economy capable of fully benefiting from it ? In my view, that's far from guaranteed.
xitter.gq/C_Barraud/status/21200…
🇨🇳 China's exports surged 35.2% YoY in October with high-tech exports up 66.6% and integrated circuits jumping 137.1%. Impressive numbers but a rather worrying economic paradox lies.
📊 High-tech products now account for nearly 34% of China's total exports and explain roughly 52% of their annual increase in value. China is benefiting massively from the global technology investment cycle and specifically in AI. However, Chinese exports of integrated circuits increased by 111.1% in value over the first ten months of the year but by only 3.2% in volume. This divergence reflects an increase in the average value of exported components rather than a comparable surge in the quantities shipped.
📈 Another interesting figure is the 41.6% increase in imports. Some might interpret this as evidence of a real recovery in Chinese consumption but a significant share of these imports is used to supply China's manufacturing sector, especially its technology supply chains. Importing more components to manufacture equipment that will be exported doesn't mean domestic demand is recovering. Domestic consumption remains insufficient to sustain growth on its own and China's cumulative trade surplus has already exceeded $1 trillion over the first ten months of the year, highlighting just how dependent the country remains on foreign demand.
⚠️ This situation could become problematic in 2027. The more China benefits from global AI investment, the more certain segments of its industrial economy become dependent on that spending continuing. If hyperscalers begin slowing their capital expenditure because returns fail to justify the investments, the consequences could spread throughout the entire industrial supply chain, from electronic components to computing equipment. Add to that increasingly demanding base effects and the possibility of escalating trade tensions with the US and Europe. China is becoming simultaneously more powerful industrially and potentially more vulnerable to economic decisions made abroad.
The more China strengthens its global manufacturing dominance, the more its economic growth risks becoming dependent on external demand that it cannot fully control.
Ruishan retweeted
🌍🛰️ Supervised Semantic #Segmentation of #Urban Area Using #SAR
✍️ Joanna Pluto-Kossakowska and Sandhi Wangiyana
🔗 brnw.ch/21x77ZS




































