Search for future landing site and mining locations on Mars
Hadarou Sare
Introduction
For more than decades,NASA and its partners has been developing technologies to extract precious metals and water from the Moon, Mars and other bodies in the solar system. The extraction of water is very important because it can reduce considerably the cost of launch since the water can be split into Hydrogen and Oxygen and use as propellant in space. That idea has been clearly supported recently by the US goverment and European partners with the Artemis and other projects. However, even though scientists and engineers have already developed the right technologies able to extract water and other resources from space, it is still unclear where they should land their technologies(Rover and others) and start mining the resources they are looking for.
The goal of this project is to use the package “threejs” to visualize the future best mining locations within the best future landing/mining site on Mars(water/water ice/volatiles materials mining sites). The final product is a map showing locations of mining site within the best landing site. However, before getting into that step, I used some geological and engineering criteria to determine and selected the best landing/mining sites for the future rovers. The criteria used to select the landing/mining sites are: where are located minerals indicating the presence of water/water ice?, where do we have permanent sunlight to use as source of energy for processing the water and charging the rovers? Is the topography of the site suitable for the mobility of the rover to and from the processing center?… For this project, the most important criteria I considered is the geological criteria for question of time. I choosed and explored 9 sites as training sites and then selected among those sites Lyot crater“V00823007” which seem to be the best future landing/mining site. Within Lyot crater“V00823007”, some locations show the presence of more volatiles materials than others. So the locations with abundant volatile materials are selected as future mining site and are then ploted on a Martian georeferenced globe.
Materials and methods
I choosed 9 sites as training sites (Si polar swath“I00831006”; Lyot crater“V00823007”; Cerberus fossae“V00825005”;Echus chasma“I00839002”; Sinus Meridiana west“I00849005”; Syrtis Major“I00853002”; Ophir planum“I00864002”; Marte vallis“I00868006; Eastern Meridiani”I00836002“). For each training site, I downloaded and used Raster data from the Java Mission-planning and Analysis for Remote sensing (JMARS) website. Those Raster data are Visible and infrared data in the “.TIFF” format and are combination of nine(9) or ten (10) bands according to Mars Odysee website where a description of data are well defined. Each band is reflecting data in a specific wavelength. Even though all bands in each training site were already together in a single image in the Mars Odysee website, I decided tu use the data I downloaded from JMARS website because JMARS enable users to download and and work with each band separately (i.e. band1, band2, band3, band4…band9). I then re-added all band I downloaded from JMARS together (after using an image processing algorithm in software called Erdas Imagine to correct the images) in order to create a metadata file containing all information from each band in each site.I finally used the band ratio combination to determine the presence, abundance and distribution of water, water ice, ice or volatiles materials in each site I choosed as training site and from there I was able to see Lyot crater“V00823007” can be selected as the best future landing/mining site. I finally get the GPS coordinates of some locations within the Lyot crater“V00823007” site by going into JMARS website and plot those locations on a Martian globe using the package threejs.The locations shown on the Martian gloge which are within Lyot crater site are consider to be the future mining site because volatile materials are abundants in those locations compare to other location within Lyot crater“V00823007”.
Data and Results
PART1: Selection of the future best landing site
The goal here is to detect the site where we could have more water/ water ice/ volatiles materials. I applied some corrections to the satelitte images I downloaded. For question of time I will only show what I did in only one site here (Lyot crater) but all sites I selected as training sites were subject to the same process.
I loaded required packages
library(raster)
library(tidyverse)
library(sf)
library(rpart)
library(rpart.plot)
library(rasterVis)
library(mapedit)
library(mapview)
library(caret)
library(forcats)
library(sp)
library(raster)I prepare the data by reading in the different bands that comprise the satelitte imagery. Each band refers to different spectrum and all bands below are from Lyot crater“V00823007” image. the data I downloaded from Lyot crater can be downloaded by going to JMARS website at the location: Latitude=48, Longitude=30 with the orbit number=71153. Before downloading the data, users need to download and install JMARS. JMARS is in fact a sorte of website but specifically a software where you can find, manipulate, and download any Martian data. The url is:https://jmars.asu.edu/
Below, are the data I downloaded from JMARS for each band (in Lyot crater site). The data can be download by following the instructions given above.
library(raster)
band1 <- raster("C:/Users/Owner/Desktop/sare backup/c/user hadar/Desktop/GEO511/Tasks/geo511-tasks-hsare/geo511-2020-final-project/geo511-2020-project-hsare/selection_of_landingMining_sites/b01.tiff")
band2 <- raster("C:/Users/Owner/Desktop/sare backup/c/user hadar/Desktop/GEO511/Tasks/geo511-tasks-hsare/geo511-2020-final-project/geo511-2020-project-hsare/selection_of_landingMining_sites/b02.tiff")
band3 <- raster("C:/Users/Owner/Desktop/sare backup/c/user hadar/Desktop/GEO511/Tasks/geo511-tasks-hsare/geo511-2020-final-project/geo511-2020-project-hsare/selection_of_landingMining_sites/b03.tiff")
band4 <- raster("C:/Users/Owner/Desktop/sare backup/c/user hadar/Desktop/GEO511/Tasks/geo511-tasks-hsare/geo511-2020-final-project/geo511-2020-project-hsare/selection_of_landingMining_sites/b04.tiff")
band5 <- raster("C:/Users/Owner/Desktop/sare backup/c/user hadar/Desktop/GEO511/Tasks/geo511-tasks-hsare/geo511-2020-final-project/geo511-2020-project-hsare/selection_of_landingMining_sites/b05.tiff")
band6 <- raster("C:/Users/Owner/Desktop/sare backup/c/user hadar/Desktop/GEO511/Tasks/geo511-tasks-hsare/geo511-2020-final-project/geo511-2020-project-hsare/selection_of_landingMining_sites/b06.tiff")
band7 <- raster("C:/Users/Owner/Desktop/sare backup/c/user hadar/Desktop/GEO511/Tasks/geo511-tasks-hsare/geo511-2020-final-project/geo511-2020-project-hsare/selection_of_landingMining_sites/b07.tiff")
band8 <- raster("C:/Users/Owner/Desktop/sare backup/c/user hadar/Desktop/GEO511/Tasks/geo511-tasks-hsare/geo511-2020-final-project/geo511-2020-project-hsare/selection_of_landingMining_sites/b08.tiff")
band9 <- raster("C:/Users/Owner/Desktop/sare backup/c/user hadar/Desktop/GEO511/Tasks/geo511-tasks-hsare/geo511-2020-final-project/geo511-2020-project-hsare/selection_of_landingMining_sites/b09.tiff")Show result of band1
print(band1)## class : RasterLayer
## band : 1 (of 3 bands)
## dimensions : 3600, 320, 1152000 (nrow, ncol, ncell)
## resolution : 1, 1 (x, y)
## extent : 0, 320, 0, 3600 (xmin, xmax, ymin, ymax)
## crs : NA
## source : C:/Users/Owner/Desktop/sare backup/c/user hadar/Desktop/GEO511/Tasks/geo511-tasks-hsare/geo511-2020-final-project/geo511-2020-project-hsare/selection_of_landingMining_sites/b01.tiff
## names : b01
## values : 0, 255 (min, max)Now, I check the size of image before combining them together using STACK function to make sure the sizes match. The code used “res(bandX), where X is the number of the band” shows that all bands have the same size 20x20.
Since all bands have the same size, We could therefore add them all together to have a metadata containing all informations we are looking for and which will be helpful to continue the project. However, we add only band4, band5, band6, band8 and band9 together because those wavelength are able to reflect the distribution of water on the selected site.
#image1 <- stack(band4, band5, band6)
image2 <- stack(band4, band5, band6, band8, band9) Let’s explore the images. Here, we are able to see the number of band used, the coordinate system the imagery is projected in, the resolution (or grid cell size) of the raster.
nlayers(image2) #This give the number of bands use## [1] 5crs(image2) #the coordinate system the imagery is projected in, and the resolution (or grid cell size) of the raster.## CRS arguments: NA##Results1 Now that we know a little more about the imagery we are using, let’s plot it. Since the image is a multi-band raster, we use the plotRGB function from the raster package, which allows us to specify which bands should be visualized.
###FALSE COLOR PLOT.The false color composite uses NIR (5) for red, red (4) for green, and green (3) for blue.
par(col.axis="white",col.lab="white",tck=0)
plotRGB(image2, r = 5, g = 4, b = 3, axes = TRUE, stretch = "lin", main = "False Color Composite")
box(col="white")
###Indices of WATER
We can use Modified Normalized Difference Water Index [MNDWI=(Green-MIR)/ (Green+MIR)] to separate water from land. This idea is from Reference=https://www.researchgate.net/post/What_is_the_best_band_combination_for_highlighting_the_water_and_soil_together_in_LANDSAT_image2 The Author of the reference is Saygin Abdikan and is currently working at the Department of Geomatics Engineering, Hacettepe University. The Author researchGate is https://www.researchgate.net/profile/Saygin_Abdikan
The formula is as below: [MNDWI=(Green-MIR)/ (Green+MIR)] MNDWI <- (image[[2]] - image[[3]])/(image[[2]] + image[[3]]).
To plot the results with ggplot, we convert the raster into a data frame and use geom_tile. The plot generated by the codes below show where are located water/ water ice/ volatiles materials the most and we have then know where to land the rovers and start the mining process.
Band4, Band5, and Band6 are use for Earth datas where Band4 represent the Green band in our case and Band6 represent the MIR band.
For Mars,the formula change little bit because all bands are collected and stored in the same file in Mars Odysee website. False color or black and white could reveal the distribution of water. However, we use the formula:
MNDWI <- (band4)/(1/band9). This formula is derived after multiple simulations and observations. i.e. lot of mathematical band combinations have been tested in order to come up with the above formula.
#We derived This formula is derived after multiple observations and applied to different data before using it.
MNDWI <- (band4)/(1/band9)
#MNDWI <- (band4-band9)/(band4+band9)
as(MNDWI, "SpatialPixelsDataFrame") %>%
as.data.frame() %>%
ggplot(data = .) +
geom_tile(aes(x = x, y = y, fill = layer)) +
theme(axis.text = element_blank(),
axis.ticks = element_blank(),
panel.background = element_blank(),
panel.grid.minor = element_blank()) +
labs(title = "Water or Water ice or Volatiles materials",
x = " ",
y = " ") +
scale_fill_gradient(high = "#CEE50E",
low = "#087F28",
name = "Water or Water ice or Volatiles materials")
PART2: Visualization of mining sites within the selected landing site on Martian gloge (map)
After knowing that Lyot crater site is the best site to be consider as future landing site according to the figure above, We just get the GPS coordinates of two locations within the Lyot crater site (GPS coordinates from the onboard GPS of the spacecraft), created a “cvs” file containing the coordinates of those locations and plot them on Martian georeferenced globe using the threejs package. As state previously, the two locations within Lyot crater shown on the Martian gloge are the future mining sites.
It is important to notice that it is hard to observe a connection between the PART1 and PART2 even though both parts are connected. In fact, the mining sites appearing on the Martian globe in PART2 are from PART1. The geographical coordiantes of two potential mining site are located randomly in the figure showing “Watter or water ice or volatiles materials distribution” ploted in PART1. Those two locations according to JMARS are: 50.5 North- 29.3 East and 48.9 North-30 East. The two identified locations are then ploted in a Martian globe as final product using threej.
I loaded required packages
library(threejs)##Results2 I firstly loaded the data and then with the package called “threejs” I plot the exact locations of mining sites on the Martian global map. I created a cvs file where I stored the latitudes and logitude of the future mining sites within Lyot crater site and then call that file in R.
library(threejs)
library(csvread)
the_martian <- read.csv("C:/Users/Owner/Desktop/sare backup/c/user hadar/Desktop/GEO511/Tasks/geo511-tasks-hsare/geo511-2020-final-project/geo511-2020-project-hsare/visualize_of_landingMining_sites/the_martian.csv", comment.char="#")
#the_martian.csv <- ("C:/Users/Owner/Desktop/sare backup/c/user #hadar/Desktop/GEO511/Tasks/geo511-tasks-hsare/geo511-2020-final-project/geo511-2020-project-hsare/visualize_of_landi#ngMining_sites/the_martian.cvs")
globejs(img = "C:/Users/Owner/Desktop/sare backup/c/user hadar/Desktop/GEO511/Tasks/geo511-tasks-hsare/geo511-2020-final-project/geo511-2020-project-hsare/visualize_of_landingMining_sites/Mars_Viking_MDIM21_ClrMosaic_global_1024.jpg", bodycolor="#555555", emissive="#444444",
lightcolor="#555555",
lat = the_martian$x,
long = the_martian$y)Conclusions
The selection of the best landing site is firstly done by choosing training sites and processing satellites images to see where the water or volatiles materials are abundant. Secondly, we get the geographical coordinates of the locations consider to be best mining sites within Lyot crater and plot them on a Martian globe using the package threej. In fact, since the Martian gloge used in this project is already georeferenced with Martian coordiante system, then finding the mining site’s geographical coordinates locations allow us to plot those locations on a Martian globe using the package threejs by creating an exel file containing the coordinate of the locations and calling that exel file in Rstudio. From all sites I selected as training sites (Si polar swath“I00831006”; Lyot crater“V00823007”; Cerberus fossae“V00825005”;Echus chasma“I00839002”; Sinus Meridiana west“I00849005”; Syrtis Major“I00853002”; Ophir planum“I00864002”; Marte vallis“I00868006; Eastern Meridiani”I00836002“), Lyot crater”V00823007" seems to be the best future landing site and within Lyot crater, various locations could be potentially selected as mining site based on the results from the band ratio combination used to determine the presence, abundance and distribution of water, water ice, ice or volatiles materials within Lyot crater.
This is done using R language meaning that R is a suitable tool offering the user the choice to manipulate the data in the way he or she wants as well as to perfom diverses task in space. R is definitely a great tool that can help explorers, planetary scientists to overcome some difficulties they could not handle with other softwares.
References
Christensen, P.R., N.S. Gorelick, G.L. Mehall, and K.C. Murray, THEMIS Public Data Releases, Planetary Data System node, Arizona State University, http://themis-data.asu.edu
Christensen, P.R., B.M. Jakosky, H.H. Kieffer, M.C. Malin, H.Y. McSween, Jr., K. Nealson, G.L. Mehall, S.H. Silverman, S. Ferry, M. Caplinger, and M. Ravine, The Thermal Emission Imaging System (THEMIS) for the Mars 2001 Odyssey Mission, Space Science Reviews, 110, 85-130, 2004.
Fergason, R.L., P.R. Christensen, and H.H. Kieffer, High Resolution Thermal Inertia Derived from THEMIS: Thermal Model and Applications, J. Geophys. Res., 111, E12004, doi:10.1029/2006JE002735, 2006.
NASA/JPL-Caltech/Arizona State University
https://www.researchgate.net/profile/Saygin_Abdikan
https://cran.r-project.org/web/packages/threejs/threejs.pdf