<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Sobia Alamgir on Medium]]></title>
        <description><![CDATA[Stories by Sobia Alamgir on Medium]]></description>
        <link>https://medium.com/@sobia_alamgir?source=rss-c2580be26260------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*WZ64RiC8s6Hmx_ghtL3vHQ.png</url>
            <title>Stories by Sobia Alamgir on Medium</title>
            <link>https://medium.com/@sobia_alamgir?source=rss-c2580be26260------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Thu, 19 Nov 2026 13:15:46 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@sobia_alamgir/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Ten Days of Ice: How Freezing Rain Broke Half of Slovenia’s Forests]]></title>
            <link>https://medium.com/@sobia_alamgir/ten-days-of-ice-how-freezing-rain-broke-half-of-slovenias-forests-0a6654d29b6e?source=rss-c2580be26260------2</link>
            <guid isPermaLink="false">https://medium.com/p/0a6654d29b6e</guid>
            <category><![CDATA[slovenia]]></category>
            <category><![CDATA[science]]></category>
            <category><![CDATA[weather]]></category>
            <category><![CDATA[climate]]></category>
            <dc:creator><![CDATA[Sobia Alamgir]]></dc:creator>
            <pubDate>Wed, 18 Nov 2026 04:57:31 GMT</pubDate>
            <atom:updated>2026-11-18T04:57:31.862Z</atom:updated>
            <content:encoded><![CDATA[<p>In 2014, rain that froze on contact left 250,000 people without power. The weather that causes it is surprisingly common.</p><figure><img alt="An animated graphic titled Žled 2014. Three figures appear with blue bars: more than 50% of Slovenian forests affected, 9.3 million cubic metres of wood to fell and clear, and 429.4 million euros of storm damage in January and February 2014. Dates: 31 January to 10 February 2014." src="https://cdn-images-1.medium.com/max/640/1*D7BDk7CozhVng7LySv-tMA.gif" /><figcaption>Slovenia’s 2014 glaze storm in three numbers.</figcaption></figure><p>At the start of February 2014, the forests around Postojna were breaking under ice. Branches snapped one after another, and whole trees bent and gave way under the weight.</p><p>Slovenians call it žled: rain that freezes the moment it touches anything colder than zero. Between 31 January and 10 February 2014 it hit more than half of Slovenia’s forests. On trees around Postojna and Pivka the ice grew to roughly five centimetres thick. At the peak, about 250,000 people had no electricity, and the national weather service ARSO issued a red warning.</p><h3>How can it rain when it’s below freezing?</h3><p>It needs the air stacked like a sandwich. Snow forms high in the cloud, falls into a layer of warmer air and melts into rain. Just above the ground sits a shallow layer of air below zero. The drops pass through it too quickly to refreeze, so they arrive as supercooled water and turn to ice on whatever they hit: branches, power lines, cars, roads.</p><p>In Slovenia the warm, moist air usually arrives on a south-westerly wind at around 1,200 metres and above, while cold air stays trapped near the ground. The longer that set-up lasts, the thicker the ice.</p><h3>What the storm cost</h3><p>The forestry service planned to fell or clear 9.3 million cubic metres of wood. The government put the damage from all the storms between 30 January and 27 February 2014, a month dominated by the glaze, at 429.4 million euros: 214.3 million in forests and forest roads, and 80.5 million on the power grid.</p><p>The hardest-hit areas were Notranjska, the south-western edge of the Ljubljana basin, and the Cerkno and Idrija hills. Slovenia’s civil protection agency estimates that glaze this severe happens less than once in a hundred years. Moderate glaze, though, returns to these regions almost every winter.</p><h3>Can a radar see it coming?</h3><p>Not on its own. A weather radar sees drops in the air, but it can’t tell ordinary rain from rain that will freeze. The useful signal is the combination: precipitation on the radar falling over an area where the ground temperature is at or below zero, especially after a cold week when a warm front arrives from the south-west.</p><p>I run a Slovenian rain-radar site, and the full guide to black ice and glaze ice, in Slovenian, is here: <a href="https://radar-padavin.si/poledica-zled/">https://radar-padavin.si/poledica-zled/</a></p><p><em>Sources: URSZR (Slovenian civil protection), Zavod za gozdove Slovenije, Government of Slovenia damage assessment, ARSO. I wrote this story with help from AI and checked every figure against the sources.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=0a6654d29b6e" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The Radar Was Empty. Outside, the Wind Hit 212 km/h.]]></title>
            <link>https://medium.com/@sobia_alamgir/the-radar-was-empty-outside-the-wind-hit-212-km-h-a949a1d13b98?source=rss-c2580be26260------2</link>
            <guid isPermaLink="false">https://medium.com/p/a949a1d13b98</guid>
            <category><![CDATA[weather]]></category>
            <category><![CDATA[climate]]></category>
            <category><![CDATA[science]]></category>
            <category><![CDATA[slovenia]]></category>
            <dc:creator><![CDATA[Sobia Alamgir]]></dc:creator>
            <pubDate>Wed, 11 Nov 2026 09:48:26 GMT</pubDate>
            <atom:updated>2026-11-11T09:48:26.443Z</atom:updated>
            <content:encoded><![CDATA[<p>Why Slovenia’s bora can close motorways under a perfectly blue sky.</p><figure><img alt="An animated graphic: on the left an empty weather radar screen with a rotating sweep and the label radar: 0 mm/h, on the right a gust gauge rising from 0 to 212 km/h, white wind streaks pouring over the Nanos ridge. Text: The radar is empty. Outside, the wind is blowing 200 km/h." src="https://cdn-images-1.medium.com/max/640/1*-xHmR91cCGDMWZ_XUnDT2w.gif" /><figcaption>An empty radar screen, and a gust meter climbing to 212 km/h. Both can be true on the same morning.</figcaption></figure><p>Picture a winter morning in western Slovenia. You open the rain radar before leaving home. Not a single coloured spot over the coast. Clear sky, bright sun, great visibility. Then you drive towards Razdrto, and a gust shoves your car sideways across the lane.</p><p>That is the bora. Slovenians call it burja.</p><h3>A radar sees rain, not wind</h3><p>A weather radar measures what falls: raindrops, snowflakes, hail. Dry air is invisible to it. So when the bora blows out of a clear sky, the radar image stays completely empty, while the wind outside is strong enough to tip over lorries.</p><p>Meteorologists split it into two types. The clear (anticyclonic) bora comes with high pressure over central Europe: cold, dry air and no precipitation at all. The dark (cyclonic) bora comes with a low over the Adriatic and brings clouds, rain and sometimes snow. Only the second one shows up on radar.</p><h3>Where the 212 km/h came from</h3><p>Cold air from inland piles up behind the ridges of Nanos, Čaven and Trnovski gozd. When the pressure difference towards the Adriatic grows large enough, the air spills over the crest and races down the steep slopes into the Vipava valley.</p><p>In March 2011, a road-authority (DARS) measuring point on the Lozice viaduct of the H4 expressway recorded a gust of 212 km/h. The Slovenian Environment Agency (ARSO) lists it among the highest wind gusts ever measured in Slovenia. Its own stations measured “only” 110 to 135 km/h in the same episodes, because they are not on the most exposed spots.</p><h3>Gusts matter more than the average</h3><p>Forecasts usually give two numbers: average wind speed and gusts. With the bora they are far apart. On 10 March 2010, a buoy off Piran recorded a 30-minute average of 80 km/h, but a gust of 113 km/h. Gusts are what close roads.</p><p>Closures come in four stages, from caravans and tarpaulin trucks up to a full closure for every vehicle. The H4 between Nanos, Vipava and Ajdovščina closes most often.</p><h3>What to check instead of the radar</h3><ul><li>ARSO wind warnings and measured gusts (look at the gust column, not the average)</li><li>Road closures on promet.si</li><li>A white cloud “flag” sitting on the Nanos ridge, a sign the bora is already pouring over</li></ul><p>I put the full guide, with closure stages and station records, here: <a href="https://radar-padavin.si/burja/">https://radar-padavin.si/burja/</a></p><p>An empty radar is good news for your umbrella. It says nothing about the wind.</p><p><em>Facts from ARSO reports on the bora episodes of March 2010 and March 2011. I wrote this story with help from AI and checked every figure against the sources.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a949a1d13b98" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[One Millimetre of Rain Is One Litre of Water. That Changes How You Read a Radar.]]></title>
            <link>https://medium.com/@sobia_alamgir/one-millimetre-of-rain-is-one-litre-of-water-that-changes-how-you-read-a-radar-ddbe0bbe2bcf?source=rss-c2580be26260------2</link>
            <guid isPermaLink="false">https://medium.com/p/ddbe0bbe2bcf</guid>
            <category><![CDATA[science]]></category>
            <category><![CDATA[climate]]></category>
            <category><![CDATA[weather]]></category>
            <category><![CDATA[rain]]></category>
            <category><![CDATA[slovenia]]></category>
            <dc:creator><![CDATA[Sobia Alamgir]]></dc:creator>
            <pubDate>Wed, 04 Nov 2026 00:13:41 GMT</pubDate>
            <atom:updated>2026-11-04T00:13:41.093Z</atom:updated>
            <content:encoded><![CDATA[<h4>The number on a weather radar sounds tiny. Here is the simple conversion that makes it real.</h4><p>“Light rain, 1 mm per hour.” It sounds like almost nothing. A millimetre is the thickness of a credit card. Who carries an umbrella for that?</p><p>But that millimetre isn’t spread over a credit card. It’s spread over every square metre around you, and once you see it as water you could pour from a bottle, the numbers on a radar map start to mean something.</p><p>I run a rain radar site for Slovenia, <a href="https://radar-padavin.si/radarska-slika-padavin-24-ur/">radar-padavin.si</a>. I’m not a meteorologist. This is the conversion I wish someone had shown me on day one.</p><h3>1 mm = 1 litre per square metre</h3><p>Picture a box with a floor of exactly one metre by one metre. Leave it out in the rain. When the water inside is 1 mm deep, the box holds one litre.</p><p>That isn’t an approximation. A square metre is 10,000 square centimetres, and a 1 mm (0.1 cm) layer of water over it is exactly 1,000 cubic centimetres, which is one litre. It’s why weather services can measure rain in millimetres and mean litres at the same time.</p><figure><img alt="Rain falls into a 1 m × 1 m box. A counter rises from 0 to 1 mm, equal to 1 litre per square metre, and a bottle appears. It keeps rising to 5 mm, 5 litres, with five bottles, ending on the line: 5 mm/h = 5 litres per m² in one hour. Illustration, water level exaggerated." src="https://cdn-images-1.medium.com/max/1024/1*c1wTzruRySamSVW7_OUZ9w.gif" /><figcaption>An illustration; the water level is exaggerated so you can see it.</figcaption></figure><h3>Reading “mm/h” on a radar</h3><p>Radar images show rain intensity in millimetres per hour. That’s a rate, like the speed on a car’s dashboard: it tells you how much would fall if the rain kept going at that strength for a full hour.</p><ul><li><strong>1 mm/h</strong> for an hour: 1 litre on every square metre.</li><li><strong>5 mm/h</strong> for an hour: 5 litres per square metre. That’s proper rain; you’ll be wet in minutes.</li><li><strong>5 mm/h for ten minutes</strong>: about a sixth of that, roughly 0.8 litres per square metre.</li></ul><p>Short summer storms can show very high rates on the radar for just a few minutes. A big number for a short time can drop less water than a modest number that lasts all afternoon.</p><h3>Why the conversion is useful</h3><p>Once you think in litres, everyday questions get easy:</p><ul><li><strong>Your roof.</strong> A 100 m² roof in 20 mm of rain sheds about 2,000 litres. That’s what your gutters and drains have to carry.</li><li><strong>Your garden.</strong> 10 mm of rain gives a 20 m² vegetable patch about 200 litres, a lot more than most people would carry out with a watering can.</li><li><strong>Floods.</strong> When forecasts warn of 50 or 100 mm over a day, that’s 50 to 100 litres on every square metre of a whole catchment, all heading downhill.</li></ul><h3>A caution about precision</h3><p>Radar is brilliant for seeing where rain is and where it’s moving, but it estimates intensity from reflected signals, not from a bucket. Slovenia’s weather service, ARSO, notes that around 75 km from the radar the measured intensity is only accurate to within a factor of about three. Treat the colours as a guide to “how heavy, roughly”, and the rain gauges and official totals as the real measurement.</p><h3>The short version</h3><p>One millimetre of rain is one litre of water on every square metre. Next time the radar says 5 mm/h, picture five one-litre bottles emptied over each square metre around you, every hour it lasts.</p><p>The live radar for Slovenia, with the rain rate in mm/h, is on <a href="https://radar-padavin.si/">Radar Padavin</a>. Radar Padavin is not an official weather service. For warnings, follow ARSO.</p><p><em>I wrote this with help from an AI assistant (Claude), which checked the facts and the sources with me; any mistakes are mine.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ddbe0bbe2bcf" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Nine Degrees Apart in One Small Country: Why Slovenia’s Temperatures Differ So Much]]></title>
            <link>https://medium.com/@sobia_alamgir/nine-degrees-apart-in-one-small-country-why-slovenias-temperatures-differ-so-much-0c54dfbb30cd?source=rss-c2580be26260------2</link>
            <guid isPermaLink="false">https://medium.com/p/0c54dfbb30cd</guid>
            <category><![CDATA[science]]></category>
            <category><![CDATA[slovenia]]></category>
            <category><![CDATA[europe]]></category>
            <category><![CDATA[weather]]></category>
            <category><![CDATA[climate]]></category>
            <dc:creator><![CDATA[Sobia Alamgir]]></dc:creator>
            <pubDate>Thu, 29 Oct 2026 08:59:05 GMT</pubDate>
            <atom:updated>2026-10-29T08:59:05.303Z</atom:updated>
            <content:encoded><![CDATA[<h4>At 9:30 this morning, Portorož was at 17 °C and Škofja Loka at under 8 °C. Here’s what makes a small country so uneven.</h4><p>At 9:30 on Thursday morning, 29 October, the warmest place in my list of 29 Slovenian towns was Portorož, on the coast, at 17.1 °C. The coldest was Škofja Loka, west of Ljubljana, at 7.7 °C.</p><p>That’s a gap of 9.4 degrees in a country small enough to drive across in an afternoon. Same morning, same weather system, two very different jackets.</p><p>I run a weather radar site for Slovenia, <a href="https://radar-padavin.si/temperature-po-sloveniji/">radar-padavin.si</a>. I’m not a meteorologist. This week I added a page that lists the current temperature for 29 towns, sorted from warmest to coldest and refreshed every 15 minutes, and the first thing I noticed was how big the spread is, almost every hour of the day.</p><figure><img alt="Bars showing temperatures across Slovenia on 29 October 2026 at 9:30. Warmest: Portorož 17.1 °C, Izola 16.7 °C, Koper 16.6 °C, Nova Gorica 15.1 °C. Coldest: Celje 8.9 °C, Domžale 8.5 °C, Medvode 8.4 °C, Škofja Loka 7.7 °C. A badge at the end reads: gap 9.4 °C." src="https://cdn-images-1.medium.com/max/1024/1*UNVFsDIDySyEaU_7_oUSmg.gif" /><figcaption>Model values from Open-Meteo, 29 Oct 2026, 9:30.</figcaption></figure><p>Three things explain most of it.</p><h3>1. Height</h3><p>Air gets colder as you go up. In the international standard atmosphere, the air cools by about 6.5 °C for every 1,000 metres of height.</p><p>Slovenia packs a lot of height into a small space. Koper sits at sea level. Kranjska Gora, below the Julian Alps, is at around 810 metres. Over a 30-year average (1991–2020, from the ERA5-Land climate model), Kranjska Gora’s January comes out at about −1.7 °C and its July at about 18 °C. Koper’s are about 5 °C and 24 °C.</p><h3>2. The sea</h3><p>Water warms and cools slowly, so the Adriatic acts like a storage heater for the coast in autumn and winter.</p><p>That’s why Portorož, Izola and Koper topped the list this morning, and why they top it on most winter mornings. In the same 1991–2020 averages, Koper’s January is about 5 °C warmer than Ljubljana’s (5.3 °C against 0.2 °C), even though the two are only about 80 km apart as the crow flies.</p><p>In summer the picture flips. The hottest places are usually inland lowlands in the east and southeast. Slovenia’s highest measured temperature, 40.8 °C, was recorded at Cerklje ob Krki on 8 August 2013.</p><h3>3. Basins</h3><p>This one surprises people. On clear, calm nights, cold air is heavier than warm air, so it slides downhill and collects at the bottom of basins and valleys. By morning, a town in a basin can be colder than the hills above it, often under a layer of fog.</p><p>Much of central Slovenia sits in basins like this, from Ljubljana to Škofja Loka and Medvode. It’s a big reason they were at the bottom of this morning’s list while the coast was basking. The extreme version happens in sinkholes in the mountains: a reading of −49.1 °C was taken in a sinkhole on the Komna plateau in January 2009, though it isn’t an official record, because these spots aren’t part of the national station network.</p><h3>Models and measurements</h3><p>One honest caveat. The numbers on my page come from a weather model (Open-Meteo), adjusted for each town’s height. That’s great for a quick overview, but it isn’t a thermometer reading. Slovenia’s official measurements come from the automatic weather stations of ARSO, the national weather service, and on clear nights in basins the model and the stations can disagree by a few degrees.</p><p>So if you need the official number, go to ARSO. If you want to know at a glance where it’s warmest right now, and how far apart the coast and the Alps are, that’s what the list is for.</p><h3>The short version</h3><p>Slovenia’s temperatures differ so much because of three things: height, the sea and basins. If you’re planning a day out in autumn or winter, the coast will often be the warmest place in the country, and a valley at breakfast time the coldest.</p><p>The live list, in Slovenian, is at <a href="https://radar-padavin.si/temperature-po-sloveniji/">Temperature po Sloveniji</a>. Radar Padavin is not an official weather service. For warnings and official measurements, follow ARSO.</p><p><em>I wrote this with help from an AI assistant (Claude), which checked the facts and the sources with me; any mistakes are mine.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=0c54dfbb30cd" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Your Phone Says “Moderate Thunderstorm Warning”. Here’s What Moderate Means.]]></title>
            <link>https://medium.com/@sobia_alamgir/your-phone-says-moderate-thunderstorm-warning-heres-what-moderate-means-94b9e2e80e72?source=rss-c2580be26260------2</link>
            <guid isPermaLink="false">https://medium.com/p/94b9e2e80e72</guid>
            <category><![CDATA[safety]]></category>
            <category><![CDATA[science]]></category>
            <category><![CDATA[europe]]></category>
            <category><![CDATA[weather]]></category>
            <category><![CDATA[slovenia]]></category>
            <dc:creator><![CDATA[Sobia Alamgir]]></dc:creator>
            <pubDate>Fri, 23 Oct 2026 10:37:01 GMT</pubDate>
            <atom:updated>2026-10-23T10:37:01.881Z</atom:updated>
            <content:encoded><![CDATA[<h4>In most of Europe those words come from the national weather service, through one shared system, and each one is a colour.</h4><p>A banner slides onto your lock screen: “Moderate wind warning”. Moderate sounds mild. Should you move the car, cancel the hike, or ignore it?</p><p>I run a rain-radar site for Slovenia, <a href="https://radar-padavin.si/opozorilo-glede-zmernih-neviht/">radar-padavin.si</a>. I’m not a meteorologist. While going through my site’s search data this month, I found that one of my most visited pages gets most of its traffic from people typing the exact words of these phone alerts into Google, in Slovenian: “opozorilo glede zmernih neviht”, “opozorilo glede zmernih vetrov”. They had the warning. What they didn’t have was what it meant.</p><h3>Where the alert comes from</h3><p>Google publishes the list of sources behind the weather alerts it shows. For Slovenia, the source is the national weather service, ARSO, <a href="https://developers.google.com/maps/documentation/weather/weather-alerts">through MeteoAlarm</a>. Croatia, Austria, Italy and most other European countries are listed the same way: their own national service, through MeteoAlarm.</p><p>MeteoAlarm is the joint warning system of Europe’s national weather services. It uses one <a href="https://meteoalarm.readthedocs.io/en/latest/warnings.html">colour-coded scale</a> for everyone: yellow, orange and red, so that an orange warning means the same thing in Slovenia as in Belgium.</p><h3>The word is the colour</h3><p>The alert on your phone doesn’t show the colour. It shows a word. The word is the colour:</p><ul><li><strong>Moderate = yellow.</strong> MeteoAlarm describes it as potentially dangerous weather. The phenomena forecast are not unusual, but be attentive if you’re doing anything exposed to the weather.</li><li><strong>Severe = orange.</strong> Dangerous weather. Unusual phenomena are forecast, and damage and casualties are likely.</li><li><strong>Extreme = red.</strong> Very dangerous weather. Exceptionally intense phenomena are forecast, with major damage likely.</li></ul><p>Google’s own definitions line up: moderate is a “possible threat to life or property”, severe a “significant threat”, extreme an “extraordinary threat”.</p><p>So “moderate” isn’t “nothing”. It’s the first of three warning levels. It means: be aware, and check before you go.</p><figure><img alt="What does your phone’s alert mean: a phone shows a moderate wind warning, a severe thunderstorm warning and an extreme rain warning in turn, while a legend lights up moderate as yellow, be aware; severe as orange, be prepared; extreme as red, take action." src="https://cdn-images-1.medium.com/max/1024/1*-5Lo3fhPcU4yPCojyDq90Q.gif" /><figcaption>An illustration, not a real alert.</figcaption></figure><h3>In Slovenian, the words change</h3><p>If your phone is set to Slovenian, the same three levels arrive as “zmerno”, “močno” and “izjemno”. So “opozorilo glede zmernih vetrov” is a yellow wind warning, “opozorilo glede močne nevihte” an orange thunderstorm warning, and “opozorilo glede izjemnega dežja” a red rain warning. ARSO’s own map shows the colours; your phone shows the words.</p><h3>What to check next</h3><p>The alert tells you that something is likely, roughly where, and when. It doesn’t show you where the rain or the storm is right now. For that:</p><ul><li><strong>Thunderstorms and rain:</strong> open the live radar and watch the last 30 to 60 minutes of animation, not just the newest frame.</li><li><strong>Wind:</strong> radar can’t see wind. A clear radar image is not a calm-weather forecast, so follow the warning itself.</li><li><strong>Snow, ice and fog:</strong> these are often invisible or patchy on rain radar too, so the warning matters more than the map.</li></ul><p>For the official text and the area the warning covers, go to the national weather service or MeteoAlarm.</p><h3>The short version</h3><p>Radar Padavin explains the Slovenian alerts on its <a href="https://radar-padavin.si/opozorilo-glede-zmernih-neviht/">warnings page</a> and shows the live radar beside it. Radar Padavin is not an official weather service, so for binding warnings, follow ARSO.</p><p>Moderate is yellow. Severe is orange. Extreme is red. When your phone says “moderate”, be aware and check before you set off.</p><p><em>I wrote this with help from an AI assistant (Claude), which checked the facts and the sources with me; any mistakes are mine.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=94b9e2e80e72" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The Clocks Go Back on Sunday. Your Weather Radar’s Clock Doesn’t.]]></title>
            <link>https://medium.com/@sobia_alamgir/the-clocks-go-back-on-sunday-your-weather-radars-clock-doesn-t-4c3af70941e0?source=rss-c2580be26260------2</link>
            <guid isPermaLink="false">https://medium.com/p/4c3af70941e0</guid>
            <category><![CDATA[weather]]></category>
            <category><![CDATA[science]]></category>
            <category><![CDATA[slovenia]]></category>
            <category><![CDATA[europe]]></category>
            <category><![CDATA[daylight-saving-time]]></category>
            <dc:creator><![CDATA[Sobia Alamgir]]></dc:creator>
            <pubDate>Wed, 21 Oct 2026 04:05:21 GMT</pubDate>
            <atom:updated>2026-10-21T04:05:21.797Z</atom:updated>
            <content:encoded><![CDATA[<h4>Europe switches to winter time on 25 October. Radar images are stamped in UTC, so the sum you do in your head changes overnight.</h4><p>On Sunday morning, most of Europe gets an extra hour in bed. The clocks go back, and for a day or two half the house runs on the old time and half on the new.</p><p>Your weather radar won’t notice at all. That’s the problem.</p><p>I run a rain-radar site for Slovenia, <a href="https://radar-padavin.si/kako-brati-radarsko-sliko-padavin/">radar-padavin.si</a>. I’m not a meteorologist; I build the site and spend a lot of time reading how the official weather services label their own images. The time on a radar image is the detail that trips people up most, and this weekend it changes.</p><h3>When the clocks go back in Europe in 2026</h3><p>Across the EU, the date isn’t decided country by country. EU law fixes it: summer time ends at 1:00 a.m. GMT <a href="https://eur-lex.europa.eu/eli/dir/2000/84/oj/eng">on the last Sunday in October</a>, in every member state at the same moment.</p><p>In 2026, the last Sunday in October is the 25th. In Slovenia, Croatia, Austria, Italy and the rest of central Europe, that means 03:00 summer time becomes 02:00 winter time.</p><h3>Radar images run on UTC, which never changes</h3><p>Weather services stamp their data in UTC, Coordinated Universal Time. It’s the time on the Greenwich meridian, and it doesn’t do summer time. The US National Hurricane Center says it plainly: the switch to daylight saving time <a href="https://www.nhc.noaa.gov/aboututc.shtml">does not affect UTC</a>.</p><p>There’s a good reason for that. A radar in Slovenia, a satellite over the equator and a weather station in Portugal can only be put on one map if they all use the same clock. Local time would turn that into a mess twice a year.</p><p>Slovenia’s weather service, ARSO, prints UTC on its <a href="https://meteo.arso.gov.si/met/sl/weather/observ/radar/">radar image</a> and explains the sum on the same page: official time in Slovenia is one hour ahead of UTC in winter and two hours ahead in summer. Its own example: a measurement stamped 12:00 UTC was taken at 13:00 in winter, or 14:00 in summer.</p><h3>The sum you do in your head changes on Sunday</h3><p>Here’s what that means this week:</p><ul><li><strong>Until Saturday 24 October:</strong> 12:00 UTC on the radar = 14:00 in Slovenia. Add two hours.</li><li><strong>From Sunday 25 October:</strong> 12:00 UTC on the radar = 13:00 in Slovenia. Add one hour.</li></ul><p>The radar image looks exactly the same on Saturday and on Sunday. Only your side of the sum moves.</p><figure><img alt="Your radar clock changes on Sunday: a radar image stamped 12:00 UTC next to a local clock. Until Saturday 24 October the clock shows 14:00, UTC plus 2 hours; then the hands turn back one hour to 13:00, UTC plus 1 hour, from Sunday 25 October." src="https://cdn-images-1.medium.com/max/1024/1*42AcBe0X3z8vK0JicKTXXw.gif" /><figcaption>An illustration, not real radar data.</figcaption></figure><h3>What goes wrong if you keep adding two hours</h3><p>Say it’s Monday afternoon, 13:05 on your phone, and the newest radar frame says 12:00 UTC. If you still add two hours out of habit, you get 14:00. That frame seems to come from an hour in the future, which makes no sense, so you might decide the radar site is broken or the image is stuck.</p><p>It isn’t. 12:00 UTC is 13:00 now. The frame is a few minutes old, which is normal: ARSO says it takes about seven minutes from the start of a measurement until the image is online.</p><p>The opposite mistake happens in spring. On the last Sunday in March, the clocks go forward, and people who keep adding one hour think a fresh image is an hour old.</p><h3>Not every image uses the same clock</h3><p>One more thing to watch if you look at more than one radar. When I checked the composite image from DHMZ, Croatia’s weather service, its time label matched local time, the same clock as your phone. That’s my own observation, not something DHMZ states, so look at the label each time.</p><p>So on the Adriatic you can have two official radar images side by side, one in UTC and one in local time. After Sunday, the gap between their labels shrinks from two hours to one. If you switch between them, read the clock on each.</p><h3>The night the hour repeats</h3><p>Early on Sunday, the hour from 02:00 to 03:00 happens twice in local time. A radar animation stamped in local time would have two sets of frames labelled 02:15, 02:20 and so on. In UTC there’s no repeat: that double hour is simply 00:00 to 02:00 UTC, one frame after another. It’s one more reason weather services keep their clocks in UTC.</p><h3>The short version</h3><p>Radar Padavin shows the Slovenian radar with its UTC time, and its <a href="https://radar-padavin.si/kako-brati-radarsko-sliko-padavin/">guide to reading the radar</a> explains the same rule in Slovenian. Radar Padavin is not an official weather service, so for warnings, follow ARSO.</p><p>From Sunday, read “UTC” on the radar and add one hour. In March, go back to adding two.</p><p><em>I wrote this with help from an AI assistant (Claude), which checked the facts and the sources with me; any mistakes are mine.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=4c3af70941e0" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Why Your Rain Radar Is Empty on a Foggy Morning]]></title>
            <link>https://medium.com/@sobia_alamgir/why-your-rain-radar-is-empty-on-a-foggy-morning-1b811d98cb9b?source=rss-c2580be26260------2</link>
            <guid isPermaLink="false">https://medium.com/p/1b811d98cb9b</guid>
            <category><![CDATA[science]]></category>
            <category><![CDATA[fog]]></category>
            <category><![CDATA[driving]]></category>
            <category><![CDATA[weather]]></category>
            <category><![CDATA[slovenia]]></category>
            <dc:creator><![CDATA[Sobia Alamgir]]></dc:creator>
            <pubDate>Mon, 19 Oct 2026 04:44:55 GMT</pubDate>
            <atom:updated>2026-10-19T04:44:55.492Z</atom:updated>
            <content:encoded><![CDATA[<h4>You can’t see ten metres, and the radar shows nothing. Nothing is broken. Here’s why, and how to tell the evening before whether fog is coming.</h4><p>It’s seven in the morning in a Slovenian valley. The street lamp across the road is a smudge. You open the rain radar on your phone, expecting a grey blob over your town, and the map is completely clean. Not a single pixel.</p><p>I run a rain-radar site for Slovenia, <a href="https://radar-padavin.si/megla/">radar-padavin.si</a>, and fog is the one thing the radar can’t answer. I’m not a meteorologist. I build the site and spend a lot of time reading how the official services explain their own radar. So does fog show up on radar? Almost never. Here’s the reason, and what to look at instead.</p><h3>What counts as fog</h3><p>The UK Met Office defines fog as <a href="https://weather.metoffice.gov.uk/learn-about/weather/types-of-weather/fog">a cloud at ground level</a> that cuts visibility to less than 1,000 metres. Below 100 metres, it’s dense fog. Above 1,000 metres, it’s mist.</p><p>So fog really is a cloud, just one that happens to be sitting where you are. Keep that in mind, because it explains both of the radar’s problems.</p><h3>Why the radar misses it</h3><h4>The droplets are too small to see</h4><p>A weather radar sends out a pulse and listens for the echo bouncing back off drops in the air. The strength of that echo depends enormously on drop size. The American Meteorological Society’s glossary defines the radar reflectivity factor as the <a href="https://glossary.ametsoc.org/wiki/Radar_reflectivity_factor">sum of the sixth powers of the drop diameters</a> in a volume of air.</p><p>The sixth power is the whole story. Make a drop ten times wider and it sends back a million times more signal. Raindrops are big. Fog droplets are tiny. Their echo is so weak that it disappears. NOAA says it plainly on its own radar page: radar can only “see” particles within a certain size range, and a reflectivity image <a href="https://psl.noaa.gov/outreach/education/science/convection/Radar_s1.html">may not show things like fog</a>.</p><h4>The beam goes over the top</h4><p>Radars sit high up so hills don’t block them. Slovenia’s two, run by the national weather service ARSO, stand on Lisca and Pasja ravan. The beam leaves the radar at a slight upward angle, and the ground curves away beneath it, so the farther it travels, the higher above the ground it is. ARSO explains this on <a href="https://meteo.arso.gov.si/met/sl/weather/observ/radar/">its radar page</a>.</p><p>Valley fog is a shallow layer pressed against the ground. Far from the radar, the beam simply passes above it. Even if fog droplets were big enough to echo, the radar would often be looking at the clear air on top.</p><p>That’s why an empty radar image doesn’t mean a clear sky. It only means nothing big enough is falling.</p><h3>What to look at instead</h3><p>Use a satellite image. From above, valley fog shows up as a smooth, pale layer with sharp edges that follow the shape of the valleys, as if someone had poured milk into the low ground. On a clear morning it can be easy to spot from space, even when the radar shows nothing at all.</p><figure><img alt="Will there be fog tomorrow morning: an illustrated valley from 20:00 to 06:00. Under a clear sky with little wind, cold air slides down the slopes, and fog gathers between the houses on the valley floor until it fills the valley at dawn." src="https://cdn-images-1.medium.com/max/1024/1*QD2BolOtxM-OlZsZahu-Yg.gif" /><figcaption>An illustration, not measured data.</figcaption></figure><h3>How valley fog forms</h3><p>Most autumn fog in valleys is radiation fog. The Met Office describes it as fog formed by the <a href="https://weather.metoffice.gov.uk/learn-about/weather/types-of-weather/fog">cooling of land overnight</a> by thermal radiation, usually helped by clear skies and calm conditions.</p><p>During the day the ground soaks up heat. On a clear night there’s no cloud to hold that heat in, so the ground loses it to the sky and chills the thin layer of air touching it. Once that air cools to its dew point, the moisture in it can’t stay as vapour and condenses into droplets.</p><p>Valleys make it worse. Cold air is heavy, so it drains down the slopes and pools on the valley floor. That’s where the fog forms, and that’s where it stays.</p><h3>Will there be fog tomorrow morning?</h3><p>You can make a decent guess the evening before. The Met Office lists the ingredients as clear skies, light winds and plenty of moisture. On a Slovenian autumn evening, I’d check four things:</p><ul><li><strong>The sky is clearing.</strong> No cloud blanket means the ground cools fast.</li><li><strong>The wind is dropping.</strong> This one needs care. The Met Office puts it neatly: too light and you get dew; too strong and the moisture gets mixed through the air. A breath of wind is fine. A steady breeze usually stops fog.</li><li><strong>The ground is damp.</strong> Rain earlier in the day leaves plenty of moisture to work with. The Met Office also notes that fog forms more readily near lakes, rivers and reservoirs.</li><li><strong>It’s the season.</strong> The Met Office says radiation fog usually occurs in winter. In Slovenia, the foggy season starts in autumn.</li></ul><p>If all four line up and you live in a valley, plan for fog in the morning, especially if you’re driving early.</p><h3>When it clears, and when it doesn’t</h3><p>The Met Office says radiation fog usually clears soon after sunrise as the ground warms up. That’s the common case in autumn.</p><p>The same page also warns that valley fog can last for several days in calm winter weather. In December the sun is low and weak, the cold air stays trapped in the valley, and the fog can hang there all day. In winter, “it’ll be gone by lunchtime” is a hope, not a rule.</p><h3>The short version</h3><p>Does fog show up on radar? Almost never: the droplets are too small, and the beam is usually above the fog. Radar Padavin shows the Slovenian rain radar live, and its <a href="https://radar-padavin.si/megla/">fog page</a> explains the same thing in Slovenian. Radar Padavin is not an official weather service, so for warnings, follow ARSO.</p><p>If you want to know whether a valley is foggy right now, look at a satellite image. If you want to know whether it will be tomorrow, look at the evening sky.</p><p><em>I wrote this with help from an AI assistant (Claude), which checked the facts and the sources with me; any mistakes are mine.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1b811d98cb9b" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Which Rain Radar Covers the Croatian Coast? Istria, Kvarner and Dalmatia Explained]]></title>
            <link>https://medium.com/@sobia_alamgir/which-rain-radar-covers-the-croatian-coast-istria-kvarner-and-dalmatia-explained-aca66b095dc7?source=rss-c2580be26260------2</link>
            <guid isPermaLink="false">https://medium.com/p/aca66b095dc7</guid>
            <category><![CDATA[slovenia]]></category>
            <category><![CDATA[croatia]]></category>
            <category><![CDATA[travel]]></category>
            <category><![CDATA[weather]]></category>
            <category><![CDATA[sailing]]></category>
            <dc:creator><![CDATA[Sobia Alamgir]]></dc:creator>
            <pubDate>Sun, 18 Oct 2026 19:04:19 GMT</pubDate>
            <atom:updated>2026-10-18T19:04:19.308Z</atom:updated>
            <content:encoded><![CDATA[<h4>Slovenia’s radar sees Istria, Croatia’s radars see the islands, and one of the two images runs on a different clock.</h4><p>You’re on a beach in Rovinj, or anchored off Hvar, and the sky over the hills has gone the colour of wet slate. You open a Croatia weather radar on your phone. Which one, though? And does the picture show what’s happening now, or an hour ago?</p><p>I run a rain-radar site for Slovenia, <a href="https://radar-padavin.si/radarska-slika-jadran/">radar-padavin.si</a>, and the Adriatic coast is only a short drive from most of Slovenia. I’m not a meteorologist. I build and run the site, and I’ve spent a lot of time reading what the two official weather services say about their own radar images. The short version: for this coast there are two official radar images, they cover different parts of it best, and they don’t even show the same clock.</p><p>Most English pages about rain radar in Croatia give you a map and a colour key, and stop there. Here is the part they leave out.</p><h3>Croatia weather radar: the network in one paragraph</h3><p>Croatia’s national weather service, DHMZ, now runs <a href="https://meteo.hr/naslovnica_radarska-slika.php?tab=radari">six weather radars</a>: Puntijarka, Bilogora, Gradište, Goli, Debeljak and Uljenje. Three of them are new and stand on the Adriatic: Goli in Istria, Debeljak near Zadar and Uljenje on the Pelješac peninsula. They are C-band, dual-polarisation Doppler radars, so if you’ve searched for a “Croatia Doppler radar”, this is it. They came with an EU-funded project called METMONIC, which started in October 2017. The same project cut the time between scans from 15 minutes to 5. According to DHMZ, it’s the <a href="https://meteo.hr/infrastruktura.php?section=mreze_postaja&amp;param=dmm">first time all of Croatia</a>, on land and at sea, is covered by radar.</p><p>Slovenia’s weather service, ARSO, has two radars of its own, on Lisca and Pasja ravan. That matters more for the coast than you’d expect.</p><h3>Istria weather radar: open the Slovenian image</h3><p>If you’re in Istria or anywhere on the northern Adriatic, start with the <a href="https://meteo.arso.gov.si/met/sl/weather/observ/radar/">ARSO radar image</a>.</p><p>This isn’t me being loyal to my own country’s service. ARSO explains on its radar page that it chose where to cut the image so that it doesn’t slice through big geographical features. That’s why the whole of Istria and the whole Italian coast are inside the frame. A shower moving in from the Gulf of Venice or over Trieste is already on the image on its way in, before it reaches Poreč or Umag.</p><p>The Croatian composite covers Istria too, through the Goli radar. Having both is a good thing. If the two images disagree, that’s worth knowing.</p><h3>Kvarner, Dalmatia and the islands: open the Croatian image</h3><p>Go further south and the choice is simple. ARSO itself says that the Adriatic islands are covered by Croatia’s radars. So for Krk, Cres and Lošinj, for Zadar, Split and Hvar, and all the way down to Dubrovnik, open the DHMZ image.</p><p>DHMZ’s colour scale is in millimetres of rain per hour:</p><ul><li>0.1–1: negligible to light</li><li>1–5: light to moderate</li><li>5–25: moderate to heavy</li><li>25–100: heavy to very heavy</li><li>over 100: extreme, with thunderstorms</li></ul><p>A word about apps. Many radar apps don’t run radars of their own. RainViewer names DHMZ as the source of its Croatia map, and Neverin says it builds its map from national services’ images, DHMZ among them. So if two apps disagree, the official DHMZ image is the one to check.</p><p>ARSO’s scale looks different, so don’t carry one key across to the other image. On the Slovenian image, light blue is half a millimetre an hour or less. Yellow is around five, red around fifty. Violet or pink means a hundred or more, and at that colour the radar is almost always seeing hail.</p><h3>The clock trap</h3><p>This is the easiest one to get wrong.</p><p>The time printed on the ARSO image is UTC, not local time. Slovenia and Croatia are one hour ahead of UTC in winter and two hours ahead in summer. So if the ARSO image says 14:00 UTC, the rain was measured at 16:00 in summer and 15:00 in winter.</p><p>If you forget this in July, the image looks two hours old and you close it. Or worse, you think the storm you’re watching is from two hours ago, when it’s really on top of you now.</p><p>When I checked the DHMZ composite, its time label matched local time, the same clock as your phone. That’s my own observation, not something DHMZ states, so look at the label each time. If you switch between the two images, check the clock on each one.</p><h3>Even “live” radar is a few minutes old</h3><p>“Live radar” is a slight lie, and an honest one. ARSO says the measurement and its processing take about seven minutes before the image is online. So the newest ARSO frame always shows rain as it was a few minutes ago. For a slow drizzle, that doesn’t matter. For a fast storm line coming off the sea, it does.</p><figure><img alt="Which rain radar covers the Croatian coast: an illustrated map of the Adriatic coast with Istria, Kvarner and Dalmatia labelled, and a rain cell drifting in from the sea toward the shore. Labels: North: ARSO image, South: DHMZ image." src="https://cdn-images-1.medium.com/max/1024/1*pqlmYgjaNuVgtzLIArJYMw.gif" /><figcaption>An illustration, not real radar data.</figcaption></figure><h3>Storm radar for the coast: will it reach my beach?</h3><p>One frame tells you where the rain is. It doesn’t tell you where it’s going. Play the animation for the last 30 to 60 minutes and watch two things.</p><p><strong>Direction and speed.</strong> Pick the edge of the rain and see how far it moved. Here’s a worked example, for illustration only: if a shower moved 15 km in 30 minutes, it’s going at about 30 km/h, so a coast 30 km away has roughly an hour. That isn’t a typical storm speed. Storms are faster or slower, and they turn.</p><p><strong>Strength.</strong> A cell that goes from yellow to red within ten minutes is getting stronger. One that fades from red back to green is falling apart. A growing red cell heading your way is the one to take seriously, especially on a boat.</p><h3>What the radar can’t see</h3><p>Radar is very good, but it has blind spots, and on this coast you’ll meet all of them.</p><ul><li><strong>Mountains.</strong> The beam travels in a straight line, and mountains block part of it. Behind high ground, the image can show less rain than is really falling.</li><li><strong>Low drizzle and fog.</strong> The Earth curves away under the beam, so far from the radar it passes above low cloud. You can be standing in light drizzle or low fog while the image is empty.</li><li><strong>Bura.</strong> The strong, cold north-easterly can bring violent gusts with little or nothing on the radar. Radar measures falling drops, not wind. A clear radar image is not a calm-sea forecast, so for wind, follow the DHMZ warnings and the marine forecast.</li></ul><h3>Both images on one page</h3><p>Radar Padavin puts the ARSO and DHMZ images together on one page, the <a href="https://radar-padavin.si/radarska-slika-jadran/">Adriatic rain radar page</a>, with the clock and colour notes underneath. It’s in Slovenian, but the radar pictures don’t need translating.</p><p>Radar Padavin is not an official weather service. For warnings, follow DHMZ in Croatia and ARSO in Slovenia, and on a boat, follow the marine forecast. A radar image tells you where the rain is right now. It doesn’t tell you what the sea will do next.</p><p>So, which Croatia weather radar should you open? In Istria, open ARSO and remember the clock. Everywhere south of it, open DHMZ. In either case, watch the animation, not just the last frame.</p><p><em>I wrote this with help from an AI assistant (Claude), which checked the facts and the sources with me; any mistakes are mine.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=aca66b095dc7" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Why Do Bridges Freeze Before Roads? Most Explanations Get It Wrong]]></title>
            <link>https://medium.com/@sobia_alamgir/why-do-bridges-freeze-before-roads-most-explanations-get-it-wrong-096a985066c4?source=rss-c2580be26260------2</link>
            <guid isPermaLink="false">https://medium.com/p/096a985066c4</guid>
            <category><![CDATA[driving]]></category>
            <category><![CDATA[road-safety]]></category>
            <category><![CDATA[weather]]></category>
            <category><![CDATA[science]]></category>
            <category><![CDATA[winter]]></category>
            <dc:creator><![CDATA[Sobia Alamgir]]></dc:creator>
            <pubDate>Fri, 16 Oct 2026 10:37:45 GMT</pubDate>
            <atom:updated>2026-10-16T10:37:45.578Z</atom:updated>
            <content:encoded><![CDATA[<h4>It isn’t the metal, and it isn’t the water underneath. It’s the missing ground, a clear sky, and a road surface colder than the forecast.</h4><p>I run a rain-radar website for Slovenia, and every autumn the same question lands in my inbox in one form or another: why do bridges freeze before roads? It sounds like trivia until you’re the one braking on an overpass at seven in the morning and feeling the car keep going.</p><p>The answer you’ll usually find online is a list of three reasons. Two of them are wrong, and the right one is explained so vaguely that it doesn’t help you predict anything. So here’s the version I wish I’d read first: what’s really happening, why your weather app can’t warn you about it, what the warning sign actually promises, and what to do on the next icy morning.</p><h3>The explanations you’ve probably read, and what’s wrong with them</h3><p><strong>“Bridges are made of metal, and metal conducts heat.”</strong> The girders underneath often are, but the surface you drive on almost never is. On road bridges the deck is usually <a href="https://en.wikipedia.org/wiki/Girder_bridge">a poured reinforced-concrete slab</a>, even when it sits on steel beams, and concrete and asphalt are what ordinary roads are made of too. If the material were the reason, a concrete road would freeze as fast as a concrete bridge. It doesn’t.</p><p><strong>“The water underneath freezes and creates cold spots.”</strong> This one has it backwards. Water that is still liquid can’t have a surface much colder than 0 °C (32 °F). On a night when the air drops well below freezing, an unfrozen river is one of the <em>warmer</em> things around. It isn’t chilling the bridge, and bridges over dry valleys or over other roads ice up early too.</p><p><strong>“Cold air sinks under the bridge” or “the heat rises and freezes.”</strong> Cold air does pool in low places, but heat doesn’t “rise and freeze”, and air flowing under a deck is only part of the story. It’s not the part that decides which surface ices first.</p><h3>Why do bridges freeze before roads? The real reason</h3><p>It comes down to one thing a bridge doesn’t have: ground.</p><p>A road lies on a thick layer of earth. That earth has been storing heat all day, and all summer, and it gives that heat back slowly. On a cold night, warmth keeps seeping up into the asphalt from below. The surface still cools, but it has a heater underneath it.</p><p>A bridge deck is a thin slab hanging in the air, with little stored heat of its own and nothing warming it from below. As the National Weather Service puts it, bridges and overpasses are <a href="https://www.weather.gov/rnk/Measure_Icing">“surrounded on all sides by the cold air and freeze more quickly.”</a> The deck loses heat through its top face and its bottom face, and gets nothing back.</p><p>The biggest loss usually isn’t the wind. It’s the sky. Every surface sends heat upward all night; meteorologists call it <a href="https://forecast.weather.gov/glossary.php?word=radiational%20cooling">radiational cooling</a>, and the NWS notes it’s “more pronounced when you have a clear sky.” Clouds send some of that heat back down, like a blanket. On a clear, calm night there’s no blanket, and a surface with no heat supply from below cools fastest. A bridge deck is the textbook case, which is why it freezes faster than the road leading up to it.</p><p>That’s why the worst bridge ice tends to come on still, starry mornings, not stormy ones. And bridges are just the clearest case: overpasses, elevated roads and stretches the sun never reaches play by the same rules.</p><figure><img alt="Why do bridges freeze before roads: animation of a clear night in which the road on the ground stays above freezing while the bridge deck, cooled from above and below, drops below 0 °C and ices over" src="https://cdn-images-1.medium.com/max/1024/1*brq7nTksQ-adoWb84YZFIA.gif" /></figure><p><em>An illustration of a clear night, not measured data.</em></p><h3>The road surface can be colder than the forecast</h3><p>When a forecast says +2 °C (36 °F), that’s the air temperature, measured by a thermometer <a href="https://content.meteoblue.com/en/research-education/specifications/data-sources/measurements/temperature/">about two metres above the ground</a>. The pavement isn’t two metres up. That’s why road authorities run <a href="https://ops.fhwa.dot.gov/weather/resources/publications/tech_briefs/rwis.htm">Road Weather Information Systems</a> with sensors that monitor “air and pavement temperatures” separately: on a clear night, the two can disagree.</p><p>So can bridges freeze above 32 degrees? The air can be above freezing while the deck surface is below it, so yes. And “what temperature do bridges freeze at?” has a slightly annoying answer: water freezes at 0 °C (32 °F), but the number that matters is the temperature of the deck surface, and that number isn’t in your weather app. A forecast of +1 °C on a clear, windless night is a reason for caution, not reassurance.</p><h3>Black ice on bridges, even when it hasn’t rained</h3><p>People often assume ice needs rain or snow. On a bridge, it doesn’t.</p><ul><li><strong>Frost.</strong> British Columbia’s road-weather service explains it in one line: if the surface temperature <a href="https://www.th.gov.bc.ca/RWIS/dataDef.htm">drops below the dew point</a>, moisture from the air condenses onto the pavement, and below freezing that means “frost or black ice.”</li><li><strong>Refreezing.</strong> A deck that got wet in the afternoon, from rain, melting snow or salt spray, can freeze again after dark, long before the road does.</li><li><strong>Fog and drizzle.</strong> Light moisture settling on a cold deck does the same.</li></ul><p>This is black ice: not black at all, just so thin and clear that the road looks wet. That’s the trap.</p><p>It’s also something no weather radar can show you. Radar-padavin.si shows rain and snow moving across Slovenia in five-minute steps, but radar only sees drops falling through the air. A glaze of ice on a concrete deck is invisible to it. On a clear, freezing night the radar looks perfectly clean, and that’s exactly when bridges are at their worst.</p><h3>What the “Bridge Ices Before Road” sign really means</h3><p>In the US, the yellow diamond reading BRIDGE ICES BEFORE ROAD is sign W8–13 in the Manual on Uniform Traffic Control Devices, the rulebook the Federal Highway Administration publishes. The 2009 edition lists it as an option: it <a href="https://mutcd.fhwa.dot.gov/htm/2009/part2/part2c.htm">“may be used in advance of bridges”</a> and “may be removed or covered during seasons of the year when its message is not relevant.”</p><p>That word “may” matters. Road agencies decide for themselves where to post it and whether to take it down in summer. So a sign in July is a local choice, and a missing sign tells you nothing about the ice.</p><p>In Slovenia, where my site is based, the equivalent warning is road sign 1112, “Poledica” (black ice), which the winter road service <a href="https://www.uradni-list.si/files/RS_-2024-026-00765-OB~P002-0000.PDF">puts up and takes down with the season</a>. The physics doesn’t care which side of the Atlantic you’re on. The signs just disagree about how often to remind you.</p><h3>A five-point checklist for icy mornings</h3><ol><li><strong>Assume every bridge, overpass and shaded stretch is icy</strong> when it’s near or below freezing, especially after a clear night and before the sun reaches the road.</li><li><strong>Slow down before the bridge, not on it.</strong> Lose the speed on the road, where you still have grip. There’s no “safe speed” on black ice, only less speed.</li><li><strong>Keep the deck boring.</strong> No sudden braking, steering or acceleration. If the car starts to slide, ease off the pedals and steer smoothly where you want to go.</li><li><strong>Leave far more space.</strong> The UK Highway Code warns that in icy weather <a href="https://www.gov.uk/guidance/the-highway-code/driving-in-adverse-weather-conditions-226-to-237">stopping distances “can be ten times greater than on dry roads.”</a></li><li><strong>Check road conditions, not just the weather, and watch the surface.</strong> In the US, <a href="https://www.fhwa.dot.gov/trafficinfo/511what.htm">511</a> gives road conditions by phone, web and app; in Slovenia it’s <a href="https://www.promet.si/">promet.si</a>. As for how to spot black ice on the road: a patch that suddenly looks glossy while everything else is dry, or a car ahead whose tyres stop throwing spray, are <a href="https://wusa9.com/article/weather/what-is-black-ice/281-5805d3c6-6626-464a-b80f-9e7325cc739f">common hints of black ice</a>. They’re hints, not guarantees.</li></ol><h3>The short version</h3><p>So the next time someone asks why do bridges freeze before roads, the answer fits in one line. A bridge isn’t colder because of what it’s made of, or because of the river underneath. It’s colder because nothing is keeping it warm.</p><p>Radar-padavin.si will tell you where the rain is falling. The bridge, on a clear morning, is up to you. If you read Slovenian, the full guide to black ice and glaze ice is here: <a href="https://radar-padavin.si/poledica-zled/">Poledica in žled</a>.</p><p><em>I wrote this with help from an AI assistant (Claude), which checked the physics and the sources with me; any mistakes are mine.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=096a985066c4" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[My Cookie Banner Was “Installed” for Six Days. Nobody Ever Saw It.]]></title>
            <link>https://medium.com/@sobia_alamgir/my-cookie-banner-was-installed-for-six-days-nobody-ever-saw-it-f8244c83a297?source=rss-c2580be26260------2</link>
            <guid isPermaLink="false">https://medium.com/p/f8244c83a297</guid>
            <category><![CDATA[google-adsense]]></category>
            <category><![CDATA[seo]]></category>
            <category><![CDATA[gdpr]]></category>
            <category><![CDATA[wordpress]]></category>
            <category><![CDATA[web-development]]></category>
            <dc:creator><![CDATA[Sobia Alamgir]]></dc:creator>
            <pubDate>Mon, 12 Oct 2026 06:35:00 GMT</pubDate>
            <atom:updated>2026-10-12T06:35:00.912Z</atom:updated>
            <content:encoded><![CDATA[<h4>The plugin was active, the cache was purged, and every visitor still got a page with no banner. The outside-in checklist I use now.</h4><figure><img alt="Cookie banner not showing: a plugin marked Active next to a web page with no banner" src="https://cdn-images-1.medium.com/max/1024/1*F93Kel3yGMYhdM7aagi2uw.png" /></figure><p>I run a live rain-radar site for Slovenia, and I don’t speak Slovenian. Radar-padavin.si gets most of its visitors from Slovenia and Croatia, which means GDPR and the EU’s cookie rules apply to every single page view, and so does Google’s rule for AdSense: no certified consent banner, no personalised ads for European visitors.</p><p>So on 5 October I installed a cookie consent plugin. The plugins screen said <em>Active</em>. I moved on.</p><p>On 11 October I checked the live page properly for the first time. There had never been a banner. Not for one visitor, not for one minute. The cookie banner not showing was bad enough. What made it worse was that nothing anywhere had told me.</p><p>Here’s what went wrong, in the order I found it.</p><h3>“Installed” meant “half-configured”</h3><p>The plugin was Complianz, and it has a setup wizard. I’d clicked through it. I thought.</p><p>What I hadn’t done was pick a privacy law. The very first question, <em>which privacy law do you want to use as the default</em>, had no option selected. With no region chosen, the plugin does something perfectly logical: it decides the site doesn’t need a banner, and prints nothing.</p><p>No error. No red notice. The plugin dashboard did say something, quietly, in a list of open tasks:</p><blockquote><em>“Not all fields have been entered, or you have not clicked the ‘finish’ button yet.”</em></blockquote><p>That was the whole clue. The page source told the same story more bluntly. There was a small script from the plugin, but no banner markup at all.</p><p>It turns out this is the most common way it goes wrong. Complianz’s own <a href="https://complianz.io/cookie-banner-does-not-appear/">troubleshooting guide</a> lists “no banner is required” and an unfinished wizard as its first two causes, and a <a href="https://wordpress.org/support/?p=18561989">support thread</a> on WordPress.org describes the same silence with a different trigger. When it’s set up properly, the dashboard says so in plain words: <em>“Your site needs a cookie banner, which has been enabled.”</em> If you don’t see that sentence, you don’t have a banner.</p><h3>Cookie banner not showing? Check the page, not the dashboard</h3><p>This is the part I’d tattoo on my hand if I could. A dashboard tells you what the software thinks it’s doing. The page tells you what it’s actually doing.</p><p>The test takes two minutes, and each step has a result that proves you wrong:</p><ol><li><strong>Open the page in a private window.</strong> If you’ve ever clicked “Accept” in your normal browser, you won’t see the banner there even when it works. A private window has no stored choice.</li><li><strong>View the source and search for the banner.</strong> For Complianz, look for cmplz-cookiebanner. If the string isn&#39;t in the HTML, no amount of CSS fixing will help. The banner was never sent.</li><li><strong>Search for your analytics tag.</strong> Before consent, a working setup rewrites it so the browser won’t run it. Mine now reads type=&quot;text/plain&quot; with a data-category=&quot;statistics&quot; attribute. If your analytics script still loads normally, you&#39;re tracking people before they&#39;ve agreed, banner or no banner. That was me: Google Analytics ran for every visitor, without asking, right up to 11 October.</li></ol><p>If all three pass, the banner is live. If any one fails, the dashboard is wrong, however green it looks.</p><h3>Two caches, two purges</h3><p>Once the wizard was finished, the banner appeared. Sort of.</p><p>On a URL with a random query string tacked on (/?test=123), the banner was there. On the plain homepage, it wasn&#39;t. That&#39;s the signature of a cache serving an old copy.</p><p>I purged the WordPress cache (LiteSpeed). Most pages still showed no banner. When I looked at the response headers, the reason was sitting there in plain text: x-hcdn-cache-status: HIT. My host, Hostinger, runs its own CDN in front of WordPress, and it had kept the old pages. Worse, different edge servers held different versions. The same URL showed the banner on one request and not on the next, which made it look like the plugin was flaky.</p><p>It wasn’t. Flushing the CDN in the hosting control panel fixed it in seconds.</p><p>If you test this yourself, test the <strong>plain</strong> URL several times in a row. A cache-busting query string only proves the origin is right. Your visitors don’t use query strings.</p><p>Complianz also has a guide on <a href="https://complianz.io/javascript-delay-in-wp-rocket-and-other-caching-and-optimization-plugins/">excluding its scripts from caching and optimisation plugins</a>. Worth doing, but it solves a different problem from a stale CDN.</p><h3>The free version didn’t include the thing I installed it for</h3><p>The whole reason I needed a banner was AdSense. Since 16 January 2024, Google has required publishers serving ads in the EEA and UK to use a <a href="https://support.google.com/adsense/answer/13554116">Google-certified consent management platform (CMP) that integrates with IAB Europe’s Transparency and Consent Framework (TCF)</a>.</p><p>The plugin’s WordPress.org page talks about being Google-certified and about TCF integration. Halfway through the wizard, the TCF options sat there greyed out, labelled <em>Upgrade</em>. The free version gives you a perfectly good GDPR banner. It doesn’t give you the ads-specific mode.</p><p>That isn’t a scandal. It’s a pricing decision, and the plugin did exactly what its free tier promises. But I’d installed it for a reason it couldn’t fulfil, and I only found out inside the wizard. Before you install anything for AdSense, check the plan you’re actually on against Google’s certified list. Google’s own consent message, set up inside AdSense, is another option.</p><h3>The side effect nobody warned me about</h3><p>A consent banner that works blocks analytics until someone clicks <em>Accept</em>. Googlebot never clicks <em>Accept</em>.</p><p>If your Search Console property was verified through your Google Analytics tag, which is a common shortcut, that tag is now invisible to anyone who hasn’t consented. Google re-checks verification from time to time, so you risk losing access to your own Search Console data. My site had no other verification method in place.</p><p>The fix is boring and permanent: add Search Console’s HTML meta tag, or a DNS record, so verification doesn’t depend on a script your banner is designed to block.</p><h3>The checklist</h3><p>What I check now, before I believe any consent setup is working:</p><ol><li><strong>The wizard is finished and a region is selected.</strong> Proves you wrong if the dashboard still lists “not all fields have been entered”, or never says the site needs a banner.</li><li><strong>The banner markup is in the served HTML.</strong> Proves you wrong if a private-window view-source doesn’t contain it, or the browser console logs a “no banner” message (some plugins hide the banner from browsers that send Do Not Track).</li><li><strong>Analytics is blocked before consent.</strong> Proves you wrong if the tag isn’t rewritten to text/plain.</li><li><strong>Every cache layer is purged.</strong> That means the plugin cache <em>and</em> the host’s CDN, plus Cloudflare if you use it. Proves you wrong if the headers say HIT on a page that should have changed.</li><li><strong>The plain URL is tested several times, on several pages, including the homepage.</strong> Proves you wrong if one request out of five is missing the banner.</li><li><strong>The banner speaks your visitors’ language.</strong> Mine defaulted to English on a Slovenian site.</li><li><strong>If this is for AdSense, your plan includes certified TCF mode.</strong> Proves you wrong if the option says <em>Upgrade</em>.</li><li><strong>Search Console has its own verification.</strong> Proves you wrong if your only method is the analytics tag.</li></ol><h3>What the six days actually were</h3><p>It wasn’t a plugin bug. It was me trusting a status label. Every tool in the chain was honest about its own job. The plugin was active, the cache did what caches do, and the free tier was the free tier. Nobody’s job was to tell me the page had no banner.</p><p>Radar-padavin.si now shows the banner to every visitor, in Slovenian, and the AdSense review has been submitted. If it comes back rejected, that will be the next article.</p><p><em>I worked through this with an AI assistant (Claude), which read the plugin code and checked the live pages with me; the mistakes and the site are mine.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f8244c83a297" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>