top of page
Writer's pictureGavinOnTheMoon

Lava Flows Near Chang'e 5 Landing Site?

Updated: Mar 2, 2021

Hello everyone,


I wanted to post something with less text since recently I have not been showing you all a lot of planetary data sets I have been working with in the past couple of months. I have been working on a side project where I aim to compare radar roughness and lava margin fractal dimension data to determine whether lava flow types can be differentiated. To achieve this, I had to look for mare lava flows with margins that can be traced using ArcGIS software and has Mini-RF radar coverage. Radar remote sensing and fractal dimension data on its own is unable to differentiate most lava flow types observed on Earth. Only very smooth and very rough lava flows (e.g., smooth pahoehoe and blocky lava) can be confidently differentiated. I have looking for lunar lava flows with flow margins that are observable through visible and topographic data sets and have radar remote sensing coverage. Ensuring this remote sensing data coverage limits where I am able to find lunar lava flows. Essentially, if an area has no topographic data and radar coverage then I ignore it.


In this post, I want to show you some images of DEMs and LRO NAC stereo images of mare lava flow surfaces in the NW region of the Imbrium Mare. These lava flows are within the area, ~17 km SW from where the Chinese National Space Administration (CNSA) Chang'e 5 mission landed on December 1st. Below is one example of a DEM and LRO NAC stereo pairs I created using the Ames Stereo Pipeline tool. Funnily enough I had to remotely connect to a lab computer to remote connect to a lab computer next to it to produce this DEM. It was the only way for me to have my laptop off and let the scripts run uninterrupted.

I decided to look at the volcanic region near the Chang'e 5 landing site because it has some of the youngest volcanic deposits on the Moon and until recently it was unexplored. Using the Arizona State University (ASU) Quickmap tool I found numerous Lunar Reconnaissance Orbiter (LRO) Narrow-Angle Camera (NAC) stereo pairs around the landing site. I need stereo pairs to create LRO NAC digital elevation models (DEMs) since they show the topography of the lunar surface at the metre-scale. By adjusting the DEMs max and min elevation values, I can reveal lava flow margins that are hidden in the visible LRO NAC images.


The image below (Image 1) shows the tracings of six topographic profiles oriented perpendicular to the mare lava flow. The flow is surrounding a dome structure. The dark and bright albedo features on the slopes of the dome are a result of the illumination direction of the Sun (illuminating from the east). In the DEM image, you can see margins of a large lava flow oriented NW-SE. The flow ranges from ~3-6 km in width and unfortunately an estimated length cannot be provided because the flow is seen continuing off of the DEM. In the image, the flow is ~25 km in length. The six topographic profiles were traced by myself to show the change in elevation along the flow surface and its margins. The profiles are shown in the next image. Please follow me to the next paragraph for a description of the six profiles.

Image 1. Top: Sinusoidal map projected LRO NAC stereo pair of a mare lava flow in the Imbrium Mare near the Chang'e 5 landing site. Bottom: DEM of the mare lava flow in the LRO NAC stereo pair. Topographic profile lines (A-F) trace perpendicular to the lava flow to show the change in elevation across the surface of the flow.


I used ESRI ArcGIS to extract the elevation points along the polylines I traced. I exported the data as a .csv file (common separated value) because I wanted to see the data in a spreadsheet and not as a profile created by ArcGIS. The profile in ArcGIS was very difficult to interpret, and it was not user friendly when it came to manipulating the profile axes values and settings. Instead, I used the data in the .csv file to create my own topographic profiles through Python matplotlib. I put a hyperlink on the the matplotlib word so if you want more information I highly recommend clicking on the hyperlink and seeing the types of data plots you can create using Python. I used the following script below to make these profiles. I am still new at Python so there are probably a lot of ways I could make my script simpler and shorter. I am always looking for advice on how to improve my Python coding skills :)


Topographic Profile Python Script


%matplotlib inline

import matplotlib.pyplot as plt

plt.style.use('default')

import numpy as np

import pandas as pd


# set python to plot axes


ax = plt.axes()


# assign data in .csv file to a value


S1 = pd.read_csv(r'C:\Users\Outreach 4\Documents\Lava Flow Topography Profiles\Change 5 Volcanic Site\Change_5 Volcanic Region_A Profile_1.csv')

S2 = pd.read_csv(r'C:\Users\Outreach 4\Documents\Lava Flow Topography Profiles\Change 5 Volcanic Site\Change_5 Volcanic Region_A Profile_2.csv')

S3 = pd.read_csv(r'C:\Users\Outreach 4\Documents\Lava Flow Topography Profiles\Change 5 Volcanic Site\Change_5 Volcanic Region_A Profile_3.csv')

S4 = pd.read_csv(r'C:\Users\Outreach 4\Documents\Lava Flow Topography Profiles\Change 5 Volcanic Site\Change_5 Volcanic Region_A Profile_4.csv')

S5 = pd.read_csv(r'C:\Users\Outreach 4\Documents\Lava Flow Topography Profiles\Change 5 Volcanic Site\Change_5 Volcanic Region_A Profile_5.csv')

S6 = pd.read_csv(r'C:\Users\Outreach 4\Documents\Lava Flow Topography Profiles\Change 5 Volcanic Site\Change_5 Volcanic Region_A Profile_6.csv')


# set x and y from data in .csv file columns.


s1 = S1['Height']

s2 = S2['Height']

s3 = S3['Height']

s4 = S4['Height']

s5 = S5['Height']

s6 = S6['Height']


p1 = S1['X']

p2 = S2['X']

p3 = S3['X']

p4 = S4['X']

p5 = S5['X']

p6 = S6['X']


# Detrending profile paths. Removes increasing or decreasing trends from topographic profiles. Easier to observe and interpret fluctuations in height along the profiles.


from scipy import signal # allows you to use detrend function in Python.


s1_detrended = signal.detrend(s1)

s2_detrended = signal.detrend(s2)

s3_detrended = signal.detrend(s3)

s4_detrended = signal.detrend(s4)

s5_detrended = signal.detrend(s5)

s6_detrended = signal.detrend(s6)


# Plot detrended profiles (# x value, # y value +/- offset on y axis, # color of line) (yes, I used the American spelling for color. Sadly, Python does not accept the British spelling, colour...)


ax.plot(p1, s1_detrended - 100, color='black');

ax.plot(p2, s2_detrended - 50, color='black');

ax.plot(p3, s3_detrended, color='black');

ax.plot(p4, s4_detrended + 50, color='black');

ax.plot(p5, s5_detrended + 100, color='black');

ax.plot(p6, s6_detrended + 150, color='black');



# add axes labels and limits

plt.ylabel('Height (m)', fontsize=12)

plt.xlabel('Profile (m)', fontsize=12)

plt.ylim((130, 180))

plt.xlim((0, 8000))


plt.minorticks_on()



# adjust x and y ticks along axes based off the topographic profile data.


plt.xticks(np.arange(0, 8000, step=1000))

plt.yticks(np.arange(130, 180, step=10)



# Detrended Profile A Labels


#plt.text(7700, -85, r'Profile A', fontsize='medium')

#plt.text(100, -112, r'A', fontsize='large')

#plt.text(8200, -114, r'A`', fontsize='large')



# Detrended Profile B Labels

#plt.text(8000, -30, r'Profile B', fontsize='medium')

#plt.text(100, -60, r'B', fontsize='large')

#plt.text(9600, -65, r'B`', fontsize='large')



# Detrended Profile C labels


#plt.text(8400, 20, r'Profile C', fontsize='medium')

#plt.text(100, 0, r'C', fontsize='large')

#plt.text(9600, 0, r'C`', fontsize='large')



# Detrended Profile D Labels


#plt.text(5200, 180, r'Profile D', fontsize='medium')

#plt.text(100, 70, r'D', fontsize='large')

#plt.text(6875, 190, r'D`', fontsize='large')



# Detrended Profile E Labels


#plt.text(5700, 170, r'Profile E', fontsize='medium')

#plt.text(100, 110, r'E', fontsize='large')

#plt.text(7000, 180, r'E`', fontsize='large')



# Detrended Profile F Labels


plt.text(6700, 170, r'Profile F', fontsize='medium')

plt.text(100, 140, r'F', fontsize='large')

plt.text(7600, 146, r'F`', fontsize='large')



# add or remove grids from plots


plt.grid(b=None)


# save plot to directory. Dpi is assigned to increase image resolution.


plt.savefig('Change5 Mare Lava Flow_Detrended_Profile F.png', dpi = 1200)


# end of script



The best-fit linear trend was removed from each profile line to observe the variation in height and eliminate trends in the data. The detrend command in Python subtracts the mean from the data, removing any trends. I was considering putting all of the profiles into one plot and offsetting them, but when I tried it was difficult to see the details. I decided instead it was easy just to show you each profile in its own plot side by side. From the profiles (Image 2), you can see that A, B and C show the most variation in height on the flows surface NW of the dome structure. As we move to the SE side of the dome, profiles D and E show height variation to be less and a lot smoother. This is due to there being less craters >200 m in diameter and the topography varying a lot less (mainly green and yellow to the SE). Only when we get to profile F which is cutting across part of the dome and a few craters to do we see a greater variation in topography.

Image 2. Topographic profiles of the six profiles traced across an Imbrium mare lava flow in Image 1. The best-fit linear trend was removed from each profile by subtracting the mean from the data. These profiles were created using the Python matplotlib commands.


I want to get all of your options on the planetary images and topographic profile data I have shown you here, and tell me if you think the lava flow I have found is in fact a lava flow. It has the right morphological features categorizing it as a lava flow. However, identifying lava flows on the Moon is challenging since their surfaces have been modified by micro-meteorite gardening.


What do you all think? Is it a lava flow?


For those who like looking at planetary images and data sets, below I have added a LRO NAC DEM mosaic of the entire Chang'e landing site and volcanic region. The mosaic consists of five DEMs I generated using the Ames Stereo Pipeline.



Image 3. Mosaic of five digital elevation models of the NW region of Imbrium Mare. This region includes the Chang'e 5 Landing Site, labelled in the image. Values in the legend are in metres.


10 views0 comments

Recent Posts

See All

Comments


bottom of page