HOME PAGE

Introduction

 

ART

Introduction

Art Gallery

Photography

 

ASTRONOMY

Introduction

Choice Images

 

Jupiter 2009-15

Jupiter 2016

Jupiter 2017

Jupiter 2018

Jupiter 2019

Jup 2019 Large Pics

Jupiter 2020

Jupiter 2021

Jupiter 2022

Jupiter 2023

Jupiter 2024

Jupiter 2025

 

Saturn 2011-14

Saturn 2015

Saturn 2016

Saturn 2017

Saturn 2018

Saturn 2019

Saturn 2020

Saturn 2021

Saturn 2022

Saturn 2023

Saturn 2024

Saturn 2025

 

Mars 2010,12&14

Mars 2016

Mars 2018

Mars 2020

Mars 2022

Mars 2024

 

Uranus 2014-15

Uranus 2016-18

Uranus 2019-20

Uranus 2021-22

Uranus 2023 on

 

Neptune 2015-17

Neptune 2018-20

Neptune 2021 on

 

Sun, Moon, Venus & Miscellaneous

 

PROCESSING TUTES

 

JUPITER 2018

Note:               Some of the images on this page are "clickable" and will open up as "full scale images" in a new tab.
                        Depending on the size of your monitor you may have to click again on this image to enlarge it.
                        All the images should respond favourably to the use of the "zoom" tool of your browser page if
                        you want to up the scale further!

2nd September 2018

Last Jupiter image for 2018

Any other time I wouldn't show this image but it does have some interesting aspects; the few times we've tried to image Jove just after Sunset any other times recently the seeing was usually absolutely horrific.

Most noticeable is the darker EZ has now taken on a distinct yellowish hue to its colour.

j20180902_093230RGBpost-87685-0-61980500-1536135726

J20180902_093107RChannelpost-87685-0-23239000-1536141443

10th July 2018

In the 3 images below the darkening of the EQ Zone (EZ or Equatorial Zone) mentioned 4 images down. (when it first became visible in our own images)

j2018-07-10_11-52_rgb_dpm_700-wide

30th June 2018

j2018-06-30_10-33_rgb_dpmLarge_700-wide

j2018-06-30_11-35_rgb_dpmLarge_700-wide

18th June 2018

In the image below Jupiter first begins to display a “darkening” of the Equatorial Zone (EZ) the central region of the disk immediately below the NEB (Northern Equatorial Belt) which is the (upper) dark reddish-brown belt/band on the planet.

j2018-06-18_13-25_rgb_dpm

9th May 2018 Animations

The image below is an animation created from a series of captures taken as Jupiter approached opposition this year: the precise time of opposition in 2018 was UT 01:00 (give or take a few minutes) on May 9th 2018.

The last frame of this animation (before it reverses & sequences back to the first frame) was captured 9 hours & 10 minutes before the precise time of opposition.

Note that this is an animation using real frames created as a sequence - as opposed to those animations that rely upon a “skin” being created in WinJUPOS…the WinJUPOS types of animations look very convincing, but in my opinion have an inherent element of artificiality that I prefer to avoid! ;)

j2018-05-08-1416_9-to1550_6-REVERSING_SmootherCopy

2018-05-08-1416.9_1550.6_MAPCopy

Recent discussion about types of planetary animations on the Cloudy Nights planetary imaging forum brought up questions on the differences & merits between “traditional” animations (time lapse images arranged in a sequential, animated display) versus those created using a cylindrical projection “map” or  image” with the WinJUPOS software program such as the example above for the WinJUPOS animation here.

This WJ software, in simple terms, effectively “wraps” a map we generate from a series of images of the planet we have captured (the more the better) around a simulated globe where the viewing perspective shifts to give us the impression we are looking at a rotating globe of the planet.

2018-05-08-1416.9-JupiterCopy

Because it is only one image (albeit being made up of possibly many separate images) this type of animation can appear very impressive, with smooth, uniform detail (because it never changes) & motion that is difficult to emulate in “traditional” animations

On the other hand, the “traditional” sequential set of captured images is far more “dynamic” in that it is influenced by the varying seeing that occurs over the time taken to acquire the number of individual images used.

It is important to note that the WJ versions still utilise multiple images, but when these are combined to form the “map” that is wrapped around the “virtual globe” the single image created is a smoothed-out single “wrap” that might display differences in different sections (for instance, between the left-hand, middle & right-hand sides of said “wrap/map”) but none of these sections will vary as the virtual globe rotates – in the “traditional” animated sequence the same details will vary as the globe/disk rotates.

Once the WinJUPOS .ims files are created (& this is often done for “traditional” animated sequences as well as WJ types) the WJ animations are then created very rapidly, whilst the “traditional” still require a lot of pain-staking application!

However, I personally believe the “traditional” animation above the WJ “map” to to be a much more “real” animation because you are actually viewing a genuine rotating disk as opposed to viewing a single rotating image – but I emphasize that this is only a personal opinion.

In the example above – more images over a longer timespan would do this WJ type more justice, & remove the “phase appearance” at the start of each “rotation” - just as they would in a different manner for the comparable “traditional” example.

8th May 2018

Some detail in the GRS as well as across the disks of this image about 35 minutes apart last night & continuing the much better run of decent seeing we've been lucky enough to encounter more recently

j20180508_152318post-87685-0-11911600-1525873874_thumbPNG

5th May 2018

j2018-05-05_14-34_rgb_dpm_700-wide

4th May 2018

j2018-05-04_142254_rgb-&-j2018-04-29_142834_rgb_700-wide

29th April

Ganymede with some albedo variations in the Jovian image below

j2018-04-29_14-29_rgb_dpm-Old_700-wide

Comparison images displaying the “turbulence” in the clouds systems

before & after the GRS.in the SEB (Southern Equatorial Belt/Band) between April 29th & May 8th 2018

 

j2018-04-29to0508_14-29to15-23_rgb_dpmTILT_700-wide

25th April 2018

j2018-04-25_16-53_rgb_dpm_700-wide

2nd April 2018

Jove showed the merits of good  atmospheric conditions (transparency could've been a bit better) & delivered one of our best Jovian image so far this apparition, revealing a lot of turbulence in the SEB

j20180402_171254RGBpost-87685-0-37802300-1522916813_thumbPNG

29th March 2018

j2018-03-29_19-19_r_dpm

j2018-03-29_19-20_g_dpm

j2018-03-29_19-21_b_dpm

j2018-03-29_19-20_rgb_dpm_700-wide

20th March 2018

Jup_ut182215_RGB_200318-Drizzled-&-Non-Drizzled_Comparisons_700-wide

11th & 12th March 2018 Comparison

Whilst there is more detail in the red channel on the left below...it can be seen that it is not as "crisp" as that in the image on the right.

j20180311_12Comparisonpost-87685-0-93418400-1521247409_thumbPNG

12th March 2018

What I'd term dirty seeing in many ways due to the disk of Jove in particular appearing as if it was covered in grime...but at the same time quite a wealth of detail on said disk this phenomenon is quite familiar to us.

Here's a decent Jovian image in rgb which we were fairly pleased with for so early still

j20180312_180212RGBpost-87685-0-08242000-1521208524_thumbPNG

11th March

A quick integration of the first 3 red channels from the morning before with Io & the GRS...as noted before there's a bit of detail in the GRS, Red Channel and the RGB below it

j20180311Rpost-87685-0-13291000-1521269900_thumbPNG

j20180311_175012post-87685-0-77994800-1521202833_thumbPNG

26th February 2018

j2018-02-26_20-12_rgb_dpm_700-wide

 

 

Back to Top of Page