Hi everyone, it’s Ianjon again. I wanted to write a recap of my summer work on the Ice Cellar Monitoring project I told you about in May. I’ve done some neat things this summer. Last time I wrote a blog post I had just started on prototyping a device with a Raspberry Pi to monitor temperature and humidity for the North Slope ice cellars. Everything I’ve done I learned how to do this summer including basic python coding, soldering, and setting up a raspberry pi.
Soldering sensors to ethernet cables
There are a few iterations that I went through for the prototype, which is about 6 or 7 different versions/adaptations onto each one, making it better and more efficient for the users and cost effective . The goal of the project and this kit I’m designing is to help people across the North Slope preserve their food from hunting by providing a monitoring system that can notify them when it starts hitting a threshold on either temperature or humidity.
Testing how my prototype runs on solar power
Besides just building the prototype, I need to test if it fits in the real world, and eventually deploy these kits in ice cellars for people to use. I reached out to some of the community in June/July and was asked to present at the Alaska Eskimo Whaling Commission about the project. I presented to the Commission and they were interested in the project a fair amount. After that presentation, I was asked if I could join the ASRC-Federal STEM team on a visit to the village of Point Hope to talk about my project.
The trip to Point Hope’s main objective was to get kids of all ages interested in higher education. I followed out with ASRC-Federal employees to help with the STEM (Science, Technology, Engineering , Mathematics) program. I had never been to Point Hope before. After helping out with that, I went and interviewed some of the local people about their ice cellars and took some measurements. After the trip was over, I came back to Fairbanks to fine tune the project even further.
The summer task to design and build a prototype is fairly close to being “real -world active”. For the next year I need to run some tests in similar conditions to an ice cellar. After that’s done, fine tune some of the things and then get a real test going in various locations. I’d like to start in my hometown of Utqiagvik since it has the largest population density in the North Slope.
GINA is now producing sea surface temperature (sst) products from direct broadcast data for MODIS, AVHRR, and VIIRS. We are producing these products on a test basis to evaluate their usefulness to our users.
The satellite data is processed using the ACSPO the Community Satellite Processing Package (CSPP) release of the NOAA/STAR Advanced Clear Sky Processor for Oceans (ACSPO) processing package.
Since this product is derived from infrared imager data, areas covered by clouds are masked and assigned “no data” SST values. This is also true for land and other areas determined to have invalid SST values.
The GeoTIFF files are being produced for each pass along with a mosaic and are intended for desktop GIS users (like ArcMap). The geotiff contain a floating point temperature value in Kelvin and are made at the sensor resolution. A index of the geotiff data available can be found here.
AWIPS ready data is also being produced for use by the National Weather Service Alaska.
If this data might be useful, please drop us a line at email@example.com .
-Jay, Carl, and the rest of the folks at GINA
Typically, weather forecasters use satellites to monitor changing cloud formations, but traditional views of clouds from space are limited to the tops. At times, something much more interesting is going on below. Wouldn’t it be great if there was a satellite sensor that could see into the cloud, or even all the way through to the earth’s surface? Well in fact there is. Microwave radiation emitted from the earth and the atmosphere has a much longer wavelength than visible light or infrared radiation, so it is less affected by tiny cloud droplets.
To take advantage of this characteristic, there are currently a number of polar orbiting satellites with sensors that measure emitted microwave radiation. The information they collect is used to detect atmospheric and surface features regardless of whether they are obscured by clouds or not. Microwave sensors can detect rainfall, sea ice, soil moisture, snow cover, ocean winds and many other things.
Two microwave products are particularly important for monitoring moisture in the atmosphere. One is “Total Precipitable Water” (or TPW), and the other is “Rain Rate”. TPW is a measure of the depth of water in a column of the atmosphere if all the water in that column were precipitated as rain. Rain Rate is an estimate of the amount of rain actually falling at that instant. So the units of TPW are millimeters (or inches) and the units of rain rate are millimeters (or inches) per hour.
These two microwave products are especially important for weather forecasting in Alaska because storms that approach from the Pacific Ocean often have very little observed information that can help assess intensity. Even when these storms near the Alaskan coast, there area a limited number of weather radars and their coverage is significantly blocked by high terrain.
The animation above is a sequence of traditional 11um Infrared images from SNPP and MetOp polar satellites between 1118 UTC 17 Aug 2017 and 0447 UTC 18 Aug 2017. The color enhancement highlights colder temperatures where red to fuchsia are the coldest. Typically the coldest clouds are also the highest, and high cloud tops are assumed to have the strongest development with the most rain, but that is not always the case.
The next animation above is the microwave TPW or Total Precipitable Water product for the same time period as shown above. Here we are looking at moisture. The largest moisture values are shown in green to yellow colors. From this set of images you can see that there is a huge area of very moist air. When compared with the Infrared animation, you’ll notice this falls to the southeast of the highest cloud tops. This may indicate a potential for rain, but we still are not sure what is currently making it to the ground.
This final animation above is similar to the previous one but with Rain Rate plotted on top of TPW. In these images, pink shows lighter rain rates with blue to cyan (and a few yellow) pixels showing where the rain rates are the greatest. Maximum rain rates reach .20-.25 inches per hour in each of the passes. Although there are some areas where the cold (and thus, higher) cloud tops match the heaviest rain rates, significant rain rates extend over a much larger area than the cloud tops alone would suggest.
Also, traditional IR animation shows the clouds becoming less organized by the last frame, while significant rain rates are still indicated along the coast over central and southern southeast Alaska. Although difficult to see in this animation, one other thing to note is that the high rain rates did not extend to the northern portion of southeast Alaska.
So, what actually happened on the day of this event? Below are 24 hour rain amounts on 17 Aug 2017 for a few southeast Alaska cities. The heavier amounts were indeed significant and primarily located in the central and southern portions of the panhandle:
Southeast Alaska Rainfall August 17, 2017
Skagway 0.18 in (northern SE Alaska)
Haines 0.20 in (northern SE Alaska)
Juneau 1.20 in (central SE Alaska)
Annex Cr. 1.61 in (central SE Alaska)
Sitka 1.65 in (central SE Alaska)
Point Baker 1.11 in (central SE Alaska)
Annette 1.01 in (southern SE Alaska)
Hyder 1.48 in (southern SE Alaska)
In the previous event, the areas of high rain rates were not far from the coldest (highest) cloud tops, but as mentioned earlier this is not always the case.
In the image sequence above, the microwave rain rate is toggled on/off
as an overlay on an 11um infrared image with the same color enhancement
used previously. In this example the coldest cloud tops are well to the
east of the main rain area, more than 250 miles in some locations.
As these examples illustrate, microwave TPW and Rate Rates can be important tools for monitoring what is occurring in the whole atmosphere and not just at the top of clouds. They are especially critical for ocean storm systems where radar and other in situ observations are few and far between.
Today’s total eclipse may have missed Alaska, but weather satellites had a great view. Here is a loop from the new GOES-16 satellite showing the moon’s shadow zipping across the Lower 48.
This imagery is a combination of three channels ranging from visible light into the near infrared spectrum, and these wavelengths all have one thing in common: they respond to sunlight bouncing off of the clouds, land, and ocean. When the moon briefly blocks out the sun, such as during today’s eclipse, these is no sunlight to bounce of the targets below, and thus briefly darkness becomes visible.
GINA works with the satellite proving ground community to bring the newest tools and techniques in satellite meteorology to forecasters working with the National Weather Service and others. You may note that Alaska is not covered in this imagery, as the longitude at which GOES-16 hovers is simply too far east to allow a good look at Alaska. But worry not: GOES-17 will be in orbit within another year or two, and at a longitude far enough west to provide a great view of Alaska (and Hawaii, for that matter). Today’s imagery from GOES-16 offers Alaskans a preview of the improvements yet to come.
This loop, and the website hosting this very nice interface for interrogating all kinds of imagery from the GOES-16 satellite, is maintained by the professionals at the Cooperative Institute for Research in the Atmosphere (CIRA) in Colorado.
This week GINA team members are at Stony Brook University attending the RCN Polar Cyber Infrastructure Hackathon. Twenty attendees from national and international universities represent a variety of backgrounds and projects. Representing UAF are GINA’s technical services lead Dayne Broderson, GINA research programmer Jiang Zhu, and GINA project manager Vanessa Raymond.
For this Hackathon the team is working to create a Docker container for running Jiang’s EMODIS NDVI code on high throughput computing (HTC) using Open Science Grid. Jiang created this code for the National Park Service for annual analysis of landscape vegetation change. By using containers and HTC, the team hopes to create an efficient and cost-effective method of processing and storing this large dataset.
View our progress here:
Jiang Zhu and Dayne Broderson discussing our new approach for the hackathon
Day two: Training! Still life of training session after lunch.
Day three: Dayne Broderson presenting on our work in progress of creating and running a simple Docker container.
While the 2017 wildfire season has been comparatively quiet to date, there are indeed some fires chewing up the landscape over Alaska’s northeastern Interior, and now much of the state is beginning to notice.
This image below was taken by the Suomi National Polar Partnership (S-NPP) satellite as it glided over Alaska this afternoon. The wavelengths shown here are exactly what the human eye would see if we could hitch a ride on the S-NPP satellite. While weather satellites can detect signals across a variety of wavelengths including infrared and microwave, it turns out that the visible spectrum with which we humans are quite familiar is one of the best ways to highlight the smoke plumes from wildfires. With high pressure draped along the Brooks Range, a northeasterly flow is blowing the smoke from its point of origin in the northeastern Interior toward the southwest and across much of the state.
Here in the Fairbanks area the smoke has thus far mainly remained aloft, with the results being an orange color to the sun and a tinge of smoky smell in the air. The National Weather Service forecasts a change in wind direction after a few more days which will push the smoke eastward into Canada.