Cloudflare Radar Data Shows Total Solar Eclipse Drove Widespread Internet Traffic Drops Across Europe

Cloudflare Radar Data Shows Total Solar Eclipse Drove Widespread Internet Traffic Drops Across Europe During the August 12, 2026 total solar eclipse across Iceland, Spain, Portugal, and broader Western Europe, Cloudflare observed marked declines in Internet HTTP traffic directly aligned with the path and timing of solar obscuration. HTTP traffic volume fell between 15% and 30% along the path of totality, with regional changes overall spanning 9.3% to -46.7%. Iceland, Spain, and Portugal registered the most dramatic drops, whereas areas with minor obscuration like Denmark, Sweden, and Norway saw little change or slight increases. Network traffic mirrored obscuration levels, hitting lowest points at maximum eclipse and rebounding within minutes as people returned indoors. Cloudflare calculated traffic changes in five-minute increments against a baseline median of the prior three Wednesdays matched slot-by-slot. Traffic Decline and Totality Alignment Cloudflare Radar tracked five-minute intervals of HTTP request volume compared against a baseline of the prior three Wednesdays. As the eclipse deepened, Internet traffic dropped in tandem with solar obscuration, creating troughs that matched the moments of maximum coverage before rebounding within minutes. Countries along totality or experiencing deep partial eclipses—including Iceland, Ireland, the UK, France, Spain, and Portugal—saw traffic decline by roughly 15% to 30%, while regions with minimal obscuration showed minimal or no drop. Methodology and Regional Differences Peak solar obscuration and maximum eclipse times were derived by calculating the geometric overlap of the sun and moon every five minutes. The eclipse pathway began showing traffic shifts in Alaska around 15:35 UTC. Across the affected territories, traffic shifts ranged from 9.3% to -46.7%. While Iceland, Spain, and Portugal experienced the steepest declines, Poland and Denmark quickly stabilized, and Norway and Sweden saw minor traffic increases above baseline. Human Attention Reflected in Network Data The analysis indicates the traffic slowdowns were not driven by network failures, but directly mirrored human behavior stepping away from screens to watch the event. The downward trend across countries demonstrated that the eclipse itself was the primary cause of the temporary drop, illustrating how a physical event can collectively reshape digital activity across an entire continent.

日全食席卷欧洲引发互联网流量显著下降 Cloudflare Radar数据显示，2026年8月12日横跨冰岛、西班牙和葡萄牙等地的日全食促使大量民众走向户外，导致欧洲受影响地区的互联网流量在食甚前后出现明显低谷。 全食带覆盖区域的互联网流量下降了约15%至30%，冰岛、西班牙和葡萄牙降幅最为剧烈，各地区流量变动在9.3%至-46.7%之间。 流量下降程度与太阳被遮挡比例高度吻合，偏食较深的地区降幅显著，而遮蔽较浅的波兰、丹麦等国影响极小，挪威和瑞典甚至出现轻微上升。 流量数据以日食前连续三个周三同时间段的中位数作为基准，日食结束后数分钟内流量迅速反弹至常态。 日食路径与网络流量降幅高度一致 在2026年8月12日日全食期间，Cloudflare通过对比五分钟颗粒度的HTTP请求量发现，最大食甚时刻与流量最低谷高度重合。全食带及深偏食覆盖的冰岛、爱尔兰、英国、法国、西班牙和葡萄牙流量骤降，而太阳几乎未被遮挡的瑞典、丹麦、波兰和瑞士则几乎没有明显变化。 数据测量方法与区域差异表现 分析团队通过日月几何重叠计算了每5分钟的遮食率，并将日食当天流量与此前三个周三同时间段的中位数基准进行对比。流量变化自阿拉斯加开始显现，全食路径区域流量降幅达15%至30%。尽管受人口密度、云层覆盖等因素影响存在离散，但整体趋势表明日食是流量下降的主因。随着阳光重现，用户迅速重返屏幕，流量在食甚后数分钟内即回归正常。

At a time when looking down at our devices is a ritual in daily life, a natural phenomenon that demands our attention communally upward is a welcome event. On Wednesday, August 12, a total solar eclipse swept from the North Atlantic across Europe, moving over Iceland and northern Spain and Portugal, with a deep partial eclipse over the rest of Western Europe, all near local sunset. This was the first total solar eclipse to cross mainland Europe in twenty years, and it drew millions outdoors to witness the moon pass between Earth and the sun. As we saw [during the 2026 World Cup](https://blog.cloudflare.com/2026-world-cup-internet-traffic/) and [the last total eclipse in 2024](https://blog.cloudflare.com/total-eclipse-internet-traffic-impacts-mexico-us-canada/), online behavior is noticeably affected when an event at this scale takes place. In this blog post, we’ll use data from [Cloudflare Radar](https://radar.cloudflare.com/) to examine how Internet traffic shifted alongside the moon and the sun. In the figure above, we measured HTTP request volume in five-minute buckets across the affected countries on eclipse day, and compared it to a normal-day baseline. Each row is a country (except for Alaska) and each column is a five-minute slice of August 12, 2026, the day of the eclipse. The countries appear above in the order in which they saw the eclipse. The black diamonds mark the moment of maximum eclipse and the color shows the percent change in HTTP traffic versus the baseline (red = below normal, blue = above). We can see very clearly that the black diamonds overlay the darkest red, almost perfectly, signaling that **as the eclipse deepened, Internet traffic decreased.** The decrease was most significant along the path of totality and in countries that saw the deepest partial eclipse (Iceland, Ireland, the UK, France, Spain and Portugal) and all but absent where the sun was barely obscured (Sweden, Denmark, Poland, Switzerland). Where the eclipse was deep, the red color and black diamonds mirror each other: traffic falls into a trough that sits directly beneath the peak of obscuration and rebounds as the sun reappears, typically within minutes of maximum coverage as people return to their screens. The scatter plot above suggests that these decreases are not due to random chance. Each point represents the peak solar obscuration of an individual country (*x*-axis), against the region's traffic dip (*y*-axis). The traffic dip is measured as the average percentage change versus baseline in the 15-minute window surrounding maximum eclipse. The downward trend of the dotted line following the dots demonstrates that **regions along the path of totality saw traffic fall by roughly 15% to 30%, whereas areas experiencing only a shallow partial eclipse dipped far less or not at all.** While local variables like population density, time of day, and cloud cover account for scatter at any given coverage level, the overall direction remains consistent. Paired with the precise timing of the drops, the trend of the data demonstrates that the eclipse itself was the primary driver of the decline. The figure above shows trend lines for each individual country. The gray triangles represent the progression of the eclipse obscuration, while the red line tracks the amount of traffic changed from its usual baseline. In almost all the countries and regions impacted in the course of the eclipse, traffic made noteworthy changes ranging from 9.3 to -46.7%. The right-hand numbers on the “*y2*”-axis are the obscuration, calculated using precise sun and moon positions. For each location we found the apparent angular sizes of the sun and moon and how far apart they are in the sky, then calculated the fraction of the sun's disk covered by the moon every 5 minutes via the geometric overlap of two circles. That gave each place both its peak obscuration (how deep the eclipse got, 0–100%) and its moment of maximum eclipse. We then summed all regions in each country for the traffic total, and took the average obscuration across a country's regions to place its national max-eclipse time. To differentiate the eclipse from a typical Wednesday evening, we compared eclipse day against the same weekday: the median of the three previous Wednesdays, matched slot-by-slot on time-of-day. Using the median keeps one odd week from skewing the comparison. Every number is then reported as percent change vs. baseline. When looking at the percentage on the left-hand y-axis, 0% means "totally normal" and negative means "less active than usual." We can see the changes in traffic beginning in Alaska around 15:35 UTC, where the eclipse began its pathway. Iceland, Spain, and Portugal experienced the most dramatic drops in traffic, whereas Poland and Denmark quickly returned to pre-eclipse levels. Norway and Sweden actually saw slight traffic increases above baseline, while Denmark recorded the least overall change. Ultimately, these findings reveal a clear correlation between the path of the eclipse and human behavior online. While the severity and duration of traffic drops varied by region, Radar’s HTTP traffic data demonstrates how a shared physical event can temporarily reshape digital activity across an entire continent. Major events in the physical world remind us that digital traffic is, at its core, a direct reflection of human attention. When the moon obscured the sun across Europe, the Internet slowed down not because of network failures, but because people paused their online activity to watch. As network patterns quickly normalized post-eclipse, the data left behind offers a fascinating snapshot of how a cosmic event can momentarily realign our online world. To explore more interactive traffic insights and track how major worldwide events shape internet activity every day, visit [Cloudflare Radar](https://radar.cloudflare.com/) or follow us on social media at [@CloudflareRadar](https://twitter.com/CloudflareRadar) (X), [https://noc.social/@cloudflareradar](https://noc.social/@cloudflareradar) (Mastodon), and [@radar.cloudflare.com](https://bsky.app/profile/radar.cloudflare.com) (Bluesky).