1.0 Introduction

1.0 Introduction » 1.1 About Scatterometer and Altimeter Observations

Real-time oceanographic observations are available from weather buoys and ships carrying weather monitoring equipment. The coverage of this data is, however, very limited, particularly over the Southern Ocean. As a result, forecasters rely almost exclusively on numerical weather prediction (NWP) to forecast weather systems, such as cold fronts, and their impact on the weather and sea state. The lack of observational data also means that there are few ways to verify the accuracy of the NWP weather and marine forecasts.

Satellite-based technologies have been changing this. For example, scatterometers, such as the Advanced Scatterometer or ASCAT on the Metop satellite series, and altimeters, such as the Jason series and Saral/Altika, are providing much needed wind and wave information over the world’s marine areas, both coastal and deep ocean. This information can be integrated with ship and buoy observations and satellite imagery to better assess current ocean conditions and evaluate the accuracy of NWP model analyses and forecasts.

Plot showing significant wave height data over the North Pacific Ocean as measured by Jason-2, Cryosat-2, and Altika polar-orbiting altimeters on October 18, 2013.

If you’re not familiar with scatterometers and altimeters, they are satellite-based microwave radars that emit microwave energy towards the surface of the earth and measure the returned signal. The signals are processed to produce various geophysical parameters used in marine forecasting and other disciplines. Among the marine parameters are sea surface height, significant wave height, and wind speed from altimeters; and wind speed and direction from scatterometers.

1.0 Introduction » 1.2 About the Lesson

The case study presented in this lesson demonstrates the use of scatterometer wind and, to a lesser extent, altimeter significant wave height products in marine forecasting. The first part of the lesson reviews cold fronts and their impact on weather and sea state conditions. The second part, the case study, follows the passage of a cold front over the South Atlantic Ocean on 23 and 24 November 2013 when the Polarstern research vessel was transiting the area. Data from 25 November 2013 are used to summarize the case study and close out the event.

The case uses observational data, including MSG satellite imagery, radiosonde measurements, coded SYNOP (surface synoptic observations) reports, ship-based photographs, ASCAT scatterometer wind products, and Jason altimeter wave heights, to assess current conditions and evaluate the accuracy of wind and wave guidance from two models, the GFS and WAVEWATCH III.

Audience

The lesson is intended for operational marine forecasters, meteorologists, and meteorological technicians at coastal stations, as well as meteorology students. To get the most out of the lesson, learners should have a sound meteorological background, particularly about mid-latitude and cold front systems, and be familiar with the basic marine meteorological forecasting parameters used in South Africa. Learners should also be competent users of conventional satellite imagery, NWP model output, Skew-T sounding profiles, and other in situ observations.

It is recommended that all learners, regardless of whether or not they have used scatterometer and altimeter products before, take COMET’s 2015 lesson Using Scatterometer Wind and Altimeter Wave Estimates in Marine Forecasting before starting the case study. It explains the basics of scatterometers and altimeters, including what they are, how they work, and how to use and interpret the products. Questions related to that lesson are woven into the case.

Objectives

After going through the lesson, learners will be better prepared to use satellite scatterometer and altimeter measurements in addition to conventional observational and NWP data in marine forecast operations. More specifically, they will be able to:

  1. Describe cold fronts and their impact on sea surface conditions
  2. Describe the uses and advantages of scatterometer sea surface winds and altimeter significant wave heights in marine forecasting
  3. Interpret scatterometer wind data
  4. Use scatterometer winds and other observations to determine current weather and sea state conditions, adjusting marine NWP analyses and forecasts as needed

1.0 Introduction » 1.3 About Cold Fronts

The case study focuses on cold fronts. These are synoptic-scale swaths of cloud and precipitation associated with a significant horizontal temperature gradient. They occur in conjunction with westerly waves, depressions, or cut-off lows and should not be considered in isolation. Cold fronts occur most frequently in winter when the amplitude of westerly disturbances is greatest.

Cold fronts are part of larger systems referred to as extratropical cyclones. These storm systems range from a few hundred to a few thousand kilometers in size and occur in association with jet streams in the middle and high latitude regions of the globe both north and south of the equator (roughly 30° to 70° latitude).

MSG Day Night Colours RGB over the Southern Atlantic Ocean near South Africa on 23 Nov 2013 06 UTC

Extratropical cyclones derive their energy from the potential energy in the pole-to-equator temperature gradient, which can become concentrated within zones called fronts. In these zones, temperature changes occur rapidly, with winds abruptly shifting direction. Depending on which airmass is replaced, a cold or warm front can occur. If warm air replaces cold air, a warm front forms. If cold air replaces warm air, a cold front forms. As a cold front advances, the trailing wedge of cold air pushes warm air upward where expansion cooling causes clouds to form.

1.0 Introduction » 1.4 About Ocean Wave Dynamics and Wind Conditions

Here is a brief review of ocean wave dynamics and wind conditions that are conducive to wave growth. They are important factors to keep in mind when interpreting wind and wave height observations and model output, such as from WAVEWATCH III, which we will do in the case study.

  • Wind speed: The wind blowing on the sea surface creates waves that move in its direction. The wind flow maintains wave growth. The process continues as long as the wind speed is greater than the wave speed.
  • Fetch or fetch length: This is the distance that the wind generally blows over the sea from a constant direction and at a constant speed
  • Duration: This is how long the wind affects the fetch or fetch length

Although each of these factors is important to consider, they do not all have the same impact on wave height. Wave height is most sensitive to changes in wind speed, even when fetch or duration is limited. Therefore, when you’re looking for areas of significant wave generation, consider wind speed first, followed by fetch length and wind duration. This helps explains why accurate wind forecasts are so important to wave forecasts in marine meteorology and numerical modeling.

2.0 About the Case

2.0 About the Case » 2.1 Introducing the Polarstern

The case takes place during the Southern Hemisphere summer season, when mid-latitude weather systems are located further south of South Africa compared to the winter season.

The case follows the voyage of the Polarstern ship as it encounters a cold front. The Polarstern is a German research vessel and icebreaker that sails to and around the Antarctic area from November and March, and spends the rest of the year in the Arctic waters.

photo of German polar research vessel POLARSTERN in Atka Bay, Antarctica during supply of Neumayer-Station

The ship is equipped for many types of research, and has nine laboratories. Its meteorological observatory is run by a meteorological technician/weather observer from the German Weather Service (DWD). The observer performs routine synoptic observations and daily upper air soundings, which are supplemented by automated weather observations. In summary, the Polarstern is like a moving weather station that offers more detailed observations than regular ships and drifting weather buoys.

2.0 About the Case » 2.2 Accounts From the Polarstern

Before exploring the data, review these accounts of the conditions encountered by the Polarstern as it passed through a cold front in the South African region on 23 and 24 November 2013. The accounts will give you a better understanding of the impacts of cold fronts and the types of decisions that research and other vessels make based on them.

Said Dr. Vera Schlindwein, Chief Scientist on the Polarstern: “Until Friday evening (22 November), we spent every minute of ship time on research. But afterwards, we had to escape southwards to avoid a storm in our survey area. We found calm conditions in the centre of the storm, which allowed us to sample and image the seafloor. Back in the survey area, though, we only had 24 hours of suitable weather (until 23 November) before we were forced to go for shelter, this time northward.

According to the weather forecasters on the ship, the storm was a record system with a central pressure near 940 hPa, which reached its peak on 23 November. The forecasters realized that the ship wouldn’t be able to escape the broad storm field but there weren’t good options for avoiding it. Going north would cost the scientists severals days of research time, while going south would move the ship close to the ice edge, a risky proposition. The captain and crew chose a path that they hoped would let them avoid the worst of the storm. But the weather got worse on 24 November and the ship never reached its desired position. It was exposed to bad weather, with wind gusts up to 64kt and waves up to 10 metres. Individual rogue waves reached the ship’s bridge level (about 17m high) before the weather finally calmed down around midnight on 24 November.

Now we’ll explore the weather events more closely. We’ll start by looking at observational data for the weather and sea state, and compare the data to the model output to assess its accuracy. If the model output is consistent with the observations, we will use it to forecast the weather and sea state for the next 24 hours.

3.0 System Identification and Model Evaluation Using Observational Data

3.0 System Identification and Model Evaluation Using Observational Data » 3.1 Satellite Imagery

On 23 November 2013 at 06 UTC, the Polarstern is located at 53.1°S and 12.6°E (the yellow star on the imagery). We’ll identify current conditions by reviewing satellite imagery. Click the tabs to view the Airmass RGB and enhanced IR animations and the daytime Natural Color RGB image. Then answer the question below.

MSG Day Night Colours RGB over the Southern Atlantic Ocean near South Africa on 23 Nov 2013 06 UTC

Question

Given the cloud structures over the South Atlantic Ocean, what is the dominant weather system in the area west of the Polarstern?

A cold front is indicated on the satellite imagery. Cold fronts are depicted by cloud bands that form an "inverted comma" on satellite imagery in the Southern Hemisphere (a "comma" in the Northern Hemisphere). These cloud bands spiral into a vortex to the south of the frontal cloud band. A warm front can also be seen emanating from the vortex.

A surface front is often best depicted in visible imagery or, as in this case, the daytime Natural Color RGB where the cloud formations and positioning of the low pressure vortex are clearly visible. Notice the speckled cloud patterns south and west of the cold front that are typical within the colder and somewhat unstable airmass.

In contrast, IR imagery can be useful for identifying fronts at night. However, cloud features are not as easily identifiable due to the courser resolution of the imagery when compared to visible channels, and the more subtle temperature contrast between low clouds and ocean surface.

The Airmass RGB combines water vapour, ozone, and infrared channels to differentiate between cold and warm airmasses. While this can be useful for identifying the general location of surface fronts, detailed low-level cloud features associated directly with fronts are more difficult to see. That’s because the main purpose of the product is to highlight contrasting thermal and moisture properties of differing airmasses.

Please make a selection.

3.0 System Identification and Model Evaluation Using Observational Data » 3.2 Synoptic Report

Here is the SYNOP report for 23 November at 06 UTC, with the information decoded in the table. Review it, then answer the question below.

Synop

SMVX01 EDZW 230600 DBLK 23061 99531 30126 41498 82606 11008 21032 49985 52027 72682 885// 22282 05008 20402 331// 41209 =

Parameter

Value

Unit of measurement

Ship’s position

53.1oS, 12.6oE

Lat/Long degrees (1 decimal)

Ships heading & speed

8-Northerly @ 6-10

Cardinal points and knots

Wind speed

12

Knots

Wind direction

260

° degrees

Pressure reduced to sea level

998.5

hPa

Pressure tendency

2-Steady/unsteady increase, 2.7 hPa

hPa per 3 hours

Air Temperature

-0.8

°C

SST

-0.8

°C

Swell height

4.5

m

Swell direction

310

° degrees

Height of wind waves

1

m

Cloud type

CL6-Stratus, CM8- Ac castellanus/floccus, CH2-Ci spissatus (dense)

Cloud height

900 to 1800

Ft above ground level

Cloud cover

8

Octas

Present weather

26-Showers of snow or rain and snow in preceding hour

Past weather

8-Showers; 2-Cloud covering more than half sky throughout

Question

Based on the conditions in the SYNOP message, which statement most accurately describes the Polarstern's position with regard to the cyclone and frontal positions?

The Polarstern is within the warm sector of the storm based on the observations.

Please make a selection.

3.0 System Identification and Model Evaluation Using Observational Data » 3.3 Upper Air Data

Here’s the Skew-T from the Polarstern at 12 UTC on 22 November. The color-filled ellipses show the layers in which cloud cover can be expected. The Skew-T also shows low-level moisture from the surface up to 810 hPa and a dry layer above that. This indicates low-level cloud with little vertical extent and the likelihood of some stratocumulus formation.

Skew-T from the Polarstern Ship 22 Nov 2013 12 UTC

A photograph taken by the crew on the ship confirms the presence of some stratocumulus cloud cover at this time.

Photograph taken from the Polarstern vessel on 23 Nov 2013

3.0 System Identification and Model Evaluation Using Observational Data » 3.4 Jason-2 Significant Wave Height

3.0 System Identification and Model Evaluation Using Observational Data » 3.4 Jason-2 Significant Wave Height » 3.4.1 Introduction

When do ocean waves become hazardous to an ocean-going vessel? The height of a wave that someone considers significant and potentially hazardous depends on the size and shape of the vessel. Mariners know the physical limits of their vessels in terms of both wind speed and wave height. This allows them to reduce their risk of encountering wind and waves that exceed those limits by applying the best available analyses and forecast information on changing weather conditions and associated sea state.

Forecasting wave conditions (especially in the open ocean) is done primarily with NWP model output. Altimeters provide an important source of remotely sensed observation data, and can be used in conjunction with the model output to improve the analysis and forecast of the sea state. However, NWP models still provide forecasts for very large contiguous ocean areas. Therefore, we’ll use WAVEWATCH III data from NOAA-NCEP Environmental Modeling Centre’s Marine Modelling and Analysis Branch (MMAB) in the case since it can be configured to generate higher resolution output.

On the next page, we’ll compare significant wave height from the Jason-2 altimeter and WAVEWATCH III model. But before we do that, let’s quickly review what significant wave height (SWH) is.

SWH is the mean height of the highest one third of the waves in a wave spectrum. It is well correlated with the wave height that a skilled observer would see. The diagram shows the relationship between significant wave height and other pertinent wave characteristics.

Image graphically shows how wave height are classified

The diagram also identifies more hazardous wave characteristics, such as the highest 10% of the wave heights (H1/10), the highest 1% of the wave heights (H1/100), and a theoretical maximum wave height (HMAX), that mariners should be aware of.

3.0 System Identification and Model Evaluation Using Observational Data » 3.4 Jason-2 Significant Wave Height » 3.4.2 Compare SWH from Jason-2 and WWIII Model

Click the tabs to compare significant wave height from the Jason-2 altimeter with the WAVEWATCH III model output, then answer the questions below.

Question

How well do the significant wave height values from the Jason altimeter and WAVEWATCH III model compare at 06 UTC on 23 November? Select the option that best completes each sentence, then click Done.

The Jason altimeter shows SWH of over to the east-northeast of the ship. WAVEWATCH III shows a broad region northeast of the ship with forecast SWHs of . The measurements from both sources correlate .

The Jason altimeter shows 7 to 8 m SWH to the east-northeast of the Polarstern, while WAVEWATCH III shows a broad region to the northeast of the ship’s position with forecast SWH of over 7.5 m. This modelled SWH compares well to the altimeter SWH value of 7 m.

Note that the Polarstern reported a swell height of 4.5 m at 53.1°S, 12.6°E (at the 06 UTC synop) and a wind wave height of 1 m, for a total wave height of 5.5 m (18 ft). According to the figure above, the ship (indicated by the red dot) was at the western edge of a region of very heavy seas that were propagating eastward.

3.0 System Identification and Model Evaluation Using Observational Data » 3.5 ASCAT Wind Data

3.0 System Identification and Model Evaluation Using Observational Data » 3.5 ASCAT Wind Data » 3.5.1 Introduction

Now we’ll introduce another tool to use in your forecasting process. ASCAT (Advanced Scatterometer) wind data can help you evaluate the NWP model and potentially more accurately place weather systems.

When reading the wind barbs in the ASCAT plots, remember that they point in the direction from which the wind is blowing, and that each short barb represents 5kt, each long barb 10kt.

Metop-A ASCAT wind barbs south east of South Africa 06 UTC 23 Nov 2013

At the 06 UTC synoptic hour, the Polarstern is located at 53.1°S 12.6°E (the red star). The closest ASCAT pass is from the Metop-A satellite with an overpass time of 05:45 UTC. Both data swaths are to the east of the ship’s position.

ASCAT shows westerly winds of 10kt (5 m/s) immediately east of the ship’s position. At 06 UTC, the ship is also reporting a westerly wind but slightly stronger, at 12kt (6 m/s).

The southern sector of the scatterometer’s western data swath shows wind speeds greater than 10kt (5 m/s), with some directional variability indicated by the wind barbs. The southern sector of the eastern data swath also shows a distinct region of directional wind shear.

3.0 System Identification and Model Evaluation Using Observational Data » 3.5 ASCAT Wind Data » 3.5.2 Scatterometer Wind Information

Here's an ASCAT wind plot generated from a Metop-B overpass at 0629 UTC on 23 November. We see that the Polarstern is between the two swaths at this time.

Metop-B ASCAT wind barbs south and east of South Africa 0629 UTC 23 Nov 2013

Question 1

What wind speed and direction did the scatterometer detect at the position indicated by the red circle?

ASCAT shows westerly winds of 35kt.

Please make a selection.

Question 2

What duration do these scatterometer winds represent?

Scatterometer ocean surface winds are averaged over a large area of around 25 km. This equates to averaging over an hour in time.

Please make a selection.

3.0 System Identification and Model Evaluation Using Observational Data » 3.5 ASCAT Wind Data » 3.5.3 MLE

The Maximum Likelihood Estimate or MLE product indicates the certainty with which the ASCAT wind direction has been determined. Large MLE values indicate winds that have a greater directional variability within a grid cell (about 25 km resolution for the plots shown). For more on MLE, access the MLE section of COMET’s Using Scatterometer Wind and Altimeter Wave Estimates in Marine Forecasting lesson.

By combining scatterometer wind data with the MLE wind variability product, we see areas of wind variability (the red shading) that correspond to areas of directional shear on a smaller scale.

Metop-A ASCAT wind barbs overlaid on MLE south and east of South Africa 0545 UTC 23 Nov 2013

Metop-A ASCAT wind barbs around South Africa 0545 UTC 23 Nov 2013

3.5.4 Directional Wind Shear

There are also areas of directional wind shear (discontinuous wind direction) with larger MLE values on the 0639 UTC ASCAT-B plot.

Drawing Question

Use the drawing tools to outline areas of directional wind shear with larger MLE values on the 0639 UTC ASCAT-B plot.

Instructions: Use the pen to outline areas of directional wind shear with large MLE values, then click Done.

Tool: Tool Size: Color:

The variation in wind direction in these areas would lead to higher MLE values. Note that the MLE flags ambiguous winds regardless of the cause - they could be due to the cold front or other factors.

3.0 System Identification and Model Evaluation Using Observational Data » 3.5 ASCAT Wind Data » 3.5.5 Combined Metop-A and -B Passes

As we just saw, Metop-A and -B both pass over the area within one hour of each other, during the early morning hours of 23 November. Let's look at how the wind data compares between the two ASCAT scatterometers. View the Metop-A pass for 0545 and the Metop-B pass for 0639 UTC below. Both show westerly winds to the east of the ship's position, and a well-defined region of directional wind shear possibly related to the presence of a surface boundary or front.

If we combine the Metop-A and -B passes (completed 54 minutes apart), we see good spatial continuity in wind speed and direction within the overlapping ocean areas sampled by both satellites.

Metop-A and B ASCAT wind barbs around South Africa from both 0545 and 0639 UTC 23 Nov 2013

4.0 Forecast Future Conditions at a Projected Point

4.0 Forecast Future Conditions at a Projected Point » 4.1 Current Synoptic Report

It’s 06 UTC on 23 November. In a moment, we will forecast the position of the cold front, along with surface winds and marine meteorological parameters for 24 November at 06 UTC in the area of the ship’s projected location.

But before we do that, let’s look at the real-time weather situation using the current SYNOP report and a photograph from the ship. They will help us evaluate if the NWP model is capturing current conditions well, in which case we can have confidence in using its forecast.

Here is the SYNOP report for 12 UTC on 23 November:

Synop FM13

SMVX01 EDZW 231200 DBLK 23121 99524 30130 41194 80212 10007 20007 49950 57034 76972 877// 22282 04001 20503 329// 41208;

The present weather type (code 76972) is reported as being rain or drizzle and snow (moderate or heavy). Although the SYNOP code gives low cloud type 7 (fractostratus), the photograph from the ship shows a cloud base near and on the surface of the ocean and fairly calm sea conditions.

Photograph taken from the Polarstern vessel on 23 Nov 2013 at 12 UTC

The observed wind direction at the vessel is NNE at 12kt. The wind waves have a 5-second period at 1.5 m height. The sea swell is a WNW swell at a 12-second period, with a height of 4 m.

Given the observed weather conditions, it appears that the Polarstern may be in the area of a warm front, possibly already in the warm sector of the approaching storm system.

4.0 Forecast Future Conditions at a Projected Point » 4.2 Compare Model MSLP and Winds to Satellite Imagery

We are using the GFS model to help us analyze the Mean Sea Level Pressure (MSLP) and 10 m wind fields. We’ll compare them against the observations and predict the movement of the approaching cold front over the 24-hour forecast period beginning at 06 UTC on 23 November.

We have overlaid the GFS 10 m wind field on the airmass RGB image valid at 06 UTC on 23 November. (Use the slider to view the GFS 10 m winds on their own.) Notice how extensive the cloud band associated with the cold front is. The GFS wind fields shows that at 10 m (the surface), the cold front is actually positioned to the east and northeast of the western boundary of the cloud band. This is happening because the cloud band seen in the satellite image is actually the mid- and upper-level cloud associated with the mid- and upper-level trough systems accompanying the cold front.

beforeafter

When placing cold fronts using satellite imagery alone, it is important to keep the following points in mind:

  • Surface cold fronts are low-level systems that extend, on average, up to 700 hPa (10000 ft)
  • Not all cloud bands in the area of the suspected surface cold front are directly linked to the front itself

4.0 Forecast Future Conditions at a Projected Point » 4.3 Draw Cold Front

Drawing Question

Draw the cold front on the 10 m NWP wind field.

Instructions: Use the pen to draw the cold front, then click Done.

Tool: Tool Size: Color:

The cold front is identifiable by the boundary/zone along which we see a sharp wind shift from northwesterly to southwesterly accompanied by very strong and gusty wind speeds. Such a boundary is seen west and northwest of the Polarstern’s plotted position (the yellow star).

GFS 10m wind field overlaid on MSG airmass RGB in South Atlantic on 6 UTC 23 Nov 2013

4.0 Forecast Future Conditions at a Projected Point » 4.4 Winds and Pressure

Here we combine the MSLP and 10 m wind field to show how the winds and pressure relate to one another. Generally the strongest winds (irrespective of direction) are found in the zone where the pressure gradients are greatest. Considering that a cold front is, by definition, a zone along which there is an abrupt change in wind and pressure, this image makes sense. Also note that weaker pressure gradients are found east (ahead) of the cold front, and that stronger/greater pressure gradients are found to the west of (behind) the cold front.

beforeafter

4.0 Forecast Future Conditions at a Projected Point » 4.5 MSLP and Satellite IR

Here, we’ve overlaid the MSLP on the IR 10.8 µm image for 06 UTC on 23 November. Note the small discrepancies in how the model placed the cold front compared to the cloud band seen in the imagery. Some of this discrepancy can be the result of whether the observed cloud structures, such as cloud bands, are directly linked to the surface front. Recall from the earlier discussion that not all cloud bands in the area of a suspected cold front are directly linked to the surface front. Low-level clouds associated with a surface front may be obscured by middle and higher level clouds.

MSG IR 10.8 micrometer and GFS mlsp valid for 06 UtC on 23 Nov 2013

4.0 Forecast Future Conditions at a Projected Point » 4.6 Compare Model to Polarstern and ASCAT Data

Now we’re at the final stage where we need to evaluate whether the GFS model is suitable to be used for the forecast. The table shows that although the model is underestimating the MSLP slightly, the wind speed and direction are accurate. All of this data is valid for the Polarstern’s location at 53.1°S and 12.6°E.

Parameter

GFS

Polarstern

ASCAT

Lat/lon

53.1°S/12.6°E

53.1°S/12.6°E

53.1°S/12.6°E

Wind speed

10-15kt

12kt

not covered by satellite pass

Wind direction

NW

260°

not covered by satellite pass

MSLP

994 hPa

998.5 hPa

not covered by satellite pass

In terms placing the cold front, we can use the GFS wind plot showing the cold front for evaluation. It appears that the model is placing the cold front farther to the east than where the western boundary of the cloud band on the IR 10.8 µm image is depicting it. Earlier, we noted the possible reasons for this discrepancy.

GFS 10m wind field over South Atlantic on 6 UTC 23 Nov 2013 with location of cold front and ship's position labeled

Comparing the ASCAT and model winds and model MSLP in the tabs, it appears that the model is placing the cold front accurately. For these reasons, we can be reasonably confident in using the GFS model to forecast the likely locations, movement, weather, and marine parameters associated with the cold front for the upcoming 24-hour period.

4.0 Forecast Future Conditions at a Projected Point » 4.7 Predicted Position of Cold Front

It is 06 UTC on 23 November and we are at the final stage of the forecast process. We will use all of the data to make a 24-hour forecast of the following:

  • The position of the cold front at 06 UTC on 24 November
  • The predicted sea and atmospheric conditions at the Polarstern’s projected position at 06 UTC on 24 November

Since the GFS model has been performing well enough to be used for the short-term forecast, now we’ll use its guidance to make the prediction. We’ll give you the position of the Polarstern for the second part of the activity.

Move the slider to view the two plots. They indicate the GFS’ predicted position of the cold front as derived from the 10-m wind and MSLP fields.

beforeafter

4.0 Forecast Future Conditions at a Projected Point » 4.8 Area with Highest Wind Speeds

Drawing Question

Using the 10 m wind field valid for 06 UTC on 24 November, circle the area with the highest wind speeds.

Instructions: Use the pen to circle the area with the high wind speeds, then click Done.

Tool: Tool Size: Color:

As you can see, the strongest winds are in the region closest to the low pressure center and in the area of the cold front. Winds of 40 to 45kt are reaching severe gale force strength and would warrant the issuance of a weather warning for the area.

4.0 Forecast Future Conditions at a Projected Point » 4.9 Locations of High and Low Pressure

Drawing Question

Using the MSLP field valid for 24 November at 06 UTC, draw the locations of high and low pressure centers.

Instructions: Use the pen to draw the high and low pressure center locations, then click Done.

Tool: Tool Size: Color:

The lowest pressure is found in the area of closed pressure contours, where the values are decreasing towards the center. The high pressures are found to the west and north-east of the low pressure, respectively. High pressure cells are associated with weak pressure gradients (isobars that are farther apart) while low pressures are associated with a stronger pressure gradients (isobars that are closer together).

4.0 Forecast Future Conditions at a Projected Point » 4.10 Forecast Position of Cold Front

Now that you have looked at the wind and MLSP fields and identified the positions of strongest winds and the low pressure center, you need to forecast the likely position of the cold front at 06 UTC on 24 November.

Drawing Question

Using the 10 m wind and MLSP overlay valid for 06 UTC on 24 November, draw the cold front in its most likely position for that time.

Instructions: Use the pen to draw cold front's expected location, then click Done.

Tool: Tool Size: Color:

The cold front will likely be located in the trough extending from the low pressure cell, coincident with the strongest winds and the wind shift from northwesterly ahead of the front, turning in a counterclockwise direction across the front (from east to west).

4.0 Forecast Future Conditions at a Projected Point » 4.11 Identify Wind Speed and Direction

The Polarstern is projected to be at lat/lon 52.2°S and 13.4°E at 06 UTC on 24 November. Looking at its position (the yellow star) in relation to the position of the cold front, identify the expected wind speed and direction at that point, then answer the questions below.

GFS 10m wind overlaid on MSLP valid for 24 Nov 2013 at 6 UTC with cold front drawn and ship's position drawn for an exercise

Question

The expected wind direction is , and the expected wind speed is .

We would expect a northwesterly wind ahead of the cold front. The GFS wind and surface pressure fields show this to be the case, and predict a wind speed of about 30kt.

4.0 Forecast Future Conditions at a Projected Point » 4.12 Conditions at Polarstern

What actually happened? How well did the forecast verify? Over the 24-hour period beginning at 06 UTC 23 November, the Polarstern experienced northerly to northwesterly winds, with increasing intensity. The highest wind speeds reported from the vessel were 33kt at 15 UTC on 23 November, which decreased to 27kt at 21 UTC on 23 November before increasing again into the near gale force range. The reported wind direction is consistent with what would be expected ahead of a cold front system.

By 06 UTC on 24 November, the Polarstern was reporting a 35 to 36kt northwesterly wind, a wind wave of 4 m, and a 3 m swell wave.

4.0 Forecast Future Conditions at a Projected Point » 4.13 WW III Forecast and the Polarstern

At 06 UTC on 24 November, WAVEWATCH III was forecasting a large zone of “very rough to high” seas approaching the position of the Polarstern (the white dot) from the west. The forecast probable wave height was approaching 5.5 m (increasing to 10 to 12.5 m in the area indicated in brown). By 09 UTC, the Polarstern was expected to reach the edge of this area.

WAVEWATCH III model forecast for 06 UTC 24 November 2013 with position of Polarstern marked

At the 09 UTC synoptic hour on 24 November, the Polarstern was reporting wind waves of 4.5 m due to the increased wind activity and a swell of 3.5 m. This would lead to expected total seas of up to 8 m.

WAVEWATCH III model forecast for 09 UTC 24 November 2013

The wind speed continued to increase through the morning, and the pressure tendency showed a decreasing trend as the vessel was still approaching the front.

By 15 UTC on 24 November, the vessel was reporting severe gale force west-northwesterly winds (41 to 47kt) and wind waves of 9.5 m. These conditions compared well to the conditions forecast by WAVEWATCH III for that hour and position.

WAVEWATCH III model forecast for 15 UTC 24 November 2013 with position of Polarstern marked

In the period between the intermediate and main SYNOP reports, from 15 to 18 UTC on 24 November, there was a marked change in conditions. This leads us to speculate that this was when the vessel moved through the surface cold front. The observations that support this conclusion include the following.

  • The pressure began to increase, and continued to do so significantly well into 24 November. Before then, the pressure had shown a steady decreasing trend.
  • The wind intensity decreased from 46 to 42kt.
  • The wind direction changed to westerly, as would be expected after crossing through a frontal region.
  • While there was no precipitation at the 15 and 18 UTC synoptic observation times on 24 November, the past weather reported for both SYNOP messages reveal broken to overcast cloud conditions over the period, as well as showers in the hour preceding both synoptic observations. Such showers are expected mainly while passing through the cold front.

5.0 Closing the Case, 25 November

5.0 Closing the Case, 25 November » 5.1 Observational Data

We’ll finish the lesson by looking at the position of the cold front on 25 November and see what the weather and marine conditions were like.

The daytime Natural Color RGB image shows the position of vessel relative to the cold front. As you can see, the cold front is well to the east of the vessel, although there is still activity south of its position.

MSG Daytime Natural Color RGB showing the southern South Africa region on 25 Nov 2013 12 UTC

The photograph from the ship at 12 UTC shows mostly clear skies. The sea state is still fairly choppy, with white caps on the ocean surface. The sea swell is up to 4.5 m and the wind waves on top of that are at 2.5 m.

Photograph taken from the Polarstern vessel on 25 Nov 2013

The corresponding SYNOP report from 12 UTC on November 25 shows cumulonimbus cloud, mid-level altocumulus cloud, and upper-level cirrus cloud.

SMVX01 EDZW 251200 DBLK 25121 99523 30133 41597 62613 11009 21037 49950 52010 72782 84931 22221 04001 20605 326// 41109 =

The table summarizes the sea conditions observed at this time.

Parameter

Observation

Wind speed

13 m/s

Wind direction

W’ly

Wind waves (height)

2.5 m

Swell (height)

4.5 m

Pressure tendency

Increasing steadily

Atmospheric temperature

-0.9°C

The observations are in line with the weather and sea conditions expected behind a cold front, where pressure is rising, temperatures are colder, and sea conditions are unsettled.

5.0 Closing the Case, 25 November » 5.2 Airmass RGB and Polarstern

The Airmass RGB animation from 23 November at 06 UTC to 25 November at 12 UTC shows the movement and progression of the cold front during this period.

The table outlines the changes in in observations from on board the ship. Compare the sea conditions when the Polarstern was east of the cold front and then west of it a day later.

Parameter

06 UTC 24 November 2013

12 UTC 25 November 2013

Wind speed

18 m/s

13 m/s

Wind direction

NW’ly

W’ly

Sea level pressure

982.2 hPa

995.0 hPa

Pressure tendency

Decreasing steadily

Increasing steadily

Atmospheric temperature

2.2°C

-0.9°C

Height of wind waves

4 m

2.5 m

Height of swell

3 m

4.5 m

Weather

Precip within sight

Showers of rain/hail or both

These results fit nicely with the theory of cold fronts and the weather changes that can be expected ahead and behind them.

Looking at the evolution of the cold front, the system was well developed on 23 and 24 November, and lost its intensity by 25 November. This is seen by the structure of the cloud band associated with the cold front on the satellite imagery. It was well formed at the beginning of the period but lost its structure on 25 November after passing over the Polarstern.

The pressure rise measured by the Polarstern between 24 and 25 November is indicative of the cold front having moved over the ship’s position, with an increase of nearly 13 hPa in one day.

The wind direction also shifted from northwesterly as measured when the ship was east of the cold front, to a more westerly wind when it was west of the cold front.

Although the temperature change was not dramatic, it did decrease by about 3°C with the passage of the cold front over the Polarstern.

5.0 Closing the Case, 25 November » 5.3 ASCAT and Vis Image

The ASCAT wind data at 0645 UTC on 25 November also corresponds well with the cloud band associated with the cold front on the 06 UTC visible 0.6 um image.

ASCAT Wind Data at 0645 UTC Overlaid on MSG Vis 0.6 µm Image at 06 UTC 25 Nov 2013

ASCAT does a good job of showing the northwesterly flow ahead of the cold front. We also see the wind shift from northwesterly to southwesterly on the northwestern parts of the frontal band, and from northwesterly to westerly over southeastern portions of the frontal band. Cyclonic rotation of the winds has also been detected to the south and west of the front, in the region west and north of the storm’s vortex. In the area of cold air cumulus cloud cover to the west and south of the cold front, the winds are southwesterly, as would be expected.

5.0 Closing the Case, 25 November » 5.4 Summary

When issuing a forecast for the weather and sea conditions associated with a cold front, it is critical to use all available information - from observations ranging from traditional satellite imagery to the new types of products from scatterometers and altimeters. Having more real-time data lets you better assess the current situation (atmospheric and marine) and evaluate the performance of the NWP model fields that are being used as part of the analysis and the forecast.

The introduction of ASCAT data into marine weather forecasting lets forecasters compare wind speeds detected by the satellite against observational data and NWP wind fields. ASCAT wind data is particularly valuable in areas where there are no direct observations of surface winds.

Here is an ASCAT and MSG visible image from the case.

ASCAT Wind Data at 0645 UTC Overlaid on MSG Vis 0.6 µm Image at 06 UTC 25 Nov 2013

This next plot shows winds from ASCAT plotted over the 10 m winds forecast by the GFS model at 06 UTC on 25 November to help determine any differences between the two. Notice the remarkable similarity in the wind fields between the two datasets, especially surrounding the frontal band. The shift in wind direction from northwesterly to southwesterly across the front is consistent between both fields, as well as the slight directional perturbations in the wind field southwest of the cold front. Wind speeds between the two fields are also comparable, with the strongest winds reaching 30 to 40kt north of the storm’s center.

ASCAT Wind Data at 0645 UTC Overlaid on GFS 10 m winds at 06 UTC 25 Nov 2013

In closing, this case demonstrates how ASCAT data can be an important complimentary tool for marine weather forecasting, particularly when trying to place cold fronts and weather systems within bands of cloud cover.

6.0 Contributors

COMET Sponsors

MetEd and the COMET® Program are a part of the University Corporation for Atmospheric Research's (UCAR's) Community Programs (UCP) and are sponsored by NOAA's National Weather Service (NWS), with additional funding by:

To learn more about us, please visit the COMET website.

Project Contributors

Program Managers
  • Mark Higgins — EUMETSAT
  • Bruce Muller — UCAR/COMET
Project Leads
  • Vesa Nietosvaara — EUMETSAT
  • Marianne Weingroff — UCAR/COMET
Instructional Design
  • Marianne Weingroff — UCAR/COMET
Science Advisors
  • Patrick Dills — UCAR/COMET
  • Mbavhi Maliage — South African Weather Service (SAWS)
  • Lithakazi Shine Mkatshwa — SAWS
  • Lee-ann Simpson — SAWS
Graphics/Animations
  • Steve Deyo — UCAR/COMET
  • Marianne Weingroff — UCAR/COMET
Multimedia Authoring/Interface Design
  • Gary Pacheco — UCAR/COMET
  • Marianne Weingroff — UCAR/COMET

 

COMET Staff, Fall 2015

Director's Office
  • Dr. Rich Jeffries, Director
  • Dr. Greg Byrd, Deputy Director
  • Lili Francklyn, Business Development Specialist
Business Administration
  • Dr. Elizabeth Mulvihill Page, Group Manager
  • Lorrie Alberta, Administrator
  • Hildy Kane, Administrative Assistant
IT Services
  • Tim Alberta, Group Manager
  • Bob Bubon, Systems Administrator
  • Dolores Kiessling, Software Engineer
  • Joey Rener, Student Assistant
  • Malte Winkler, Software Engineer
Instructional and Media Services
  • Bruce Muller, Group Manager
  • Dr. Alan Bol, Scientist/Instructional Designer
  • Steve Deyo, Graphic and 3D Designer
  • Lon Goldstein, Instructional Designer
  • Bryan Guarente, Instructional Designer
  • Vickie Johnson, Instructional Designer (Casual)
  • Gary Pacheco, Web Designer and Developer
  • Sylvia Quesada, Intern
  • Sarah Ross-Lazarov, Instructional Designer (Casual)
  • Tsvetomir Ross-Lazarov, Instructional Designer
  • David Russi, Spanish Translations
  • Andrea Smith, Meteorologist/Instructional Designer
  • Marianne Weingroff, Instructional Designer
Science Group
  • Wendy Schreiber-Abshire, Group Manager
  • Paula Brown, Visiting Meteorologist, CIRA/Colorado State University
  • Dr. William Bua, Meteorologist
  • Dr. Frank Bub, Oceanographer (Casual)
  • Lis Cohen, Meteorologist (Casual)
  • Patrick Dills, Meteorologist
  • Matthew Kelsch, Hydrometeorologist
  • Dr. Elizabeth Mulvihill Page, Meteorologist
  • Amy Stevermer, Meteorologist
  • Vanessa Vincente, Visiting Meteorologist, CIRA/Colorado State University