Radar Statistical Reconnaissance ofthe 2016 InSight Landing Sites

C. Grima*, D. D. Blankenship*Contact:

cyril.grima@gmail.com*Institute for Geophysics,

University of Texas at Austin, TX 78758, USA

1. RADAR STATISTICAL RECONNAISSANCE (RSR)PRESENTATION

The NASA's InSight (Interior Exploration

using Seismic Investigations, Geodesy

and Heat Transport) Lander is scheduled

for launch and Mars landing in March and

September 2016, respectively. From 16

landing sites defined in 2012 in Elysium

Planitia [1], the selection has been

narrowed to 4 ellipses in 2014 [2].

The landing site selection process is

achieved through a combination of various

orbital remote-sensing technologies (e.g.

laser altimetry, imagery, spectro-imagery,

spectroscopy, thermal sensing) to assess

the terrain relief, surface cohesiveness,

rock height and rock abundance [3].

We present a preliminary application of

the Radar Statistical Reconnaissance

(RSR) [4, 5, 6] technique to the active 20-

MHz Shallow Radar (SHARAD)

instrument [7] to support the InSight

landing site selection. The RSR can

assess, with a single instrument, new

observables for the surface properties:

surface RMS heights, surface

permittivity, and surface heterogeneity.

References. [1] Golombek M., Redmond L. et al. (2013) “Selection of

the InSight landing site: contraints plans and progress”, LPSC XLIV,

#1691 [2] Golombek M., Warner N. et al. (2014) “Final four landing

sites for the InSight geophysical lander”, LPSC XLV, #1499. [3] Ball et

al. (2007) “Planetary and entry probes”, Cambridge University Press

[4] Grima C., Kofman W., Hérique A., Orosei R., Seu R. (2012),

“Quantitative analysis of Mars surface radar and reflectivity at 20MHz”.

Icarus 220, 84-99. [5] Grima C., Schroeder D. M., Blankenship D. D.,

and Young D., “Planetary landing zone reconnaissance using ice

penetrating radar: Concept validation in Antarctica”. Planetary and

Space Science 103, 191-204. [6] Grima C., Blankenship D. D., Young

D. A., Schroeder D. M. (2014) “Surface slope control on firn density at

Thwaites Glacier, West Antarctica: Results from airborne radar

sounding“. Geophysical Research Letters 41(19), 6787-6794. [7] Croci

R. et al. (2011) “The Shallow RADar (SHARAD) onboard the NASA

MRO mission”. Proceedings of the IEEE(99):794-807 [8] Jakeman E.

(1980), “On the statistics of K-distributed noise”, J. of Physics:

Mathematical and General, 13(1), 31–48. [9] Ulaby F. T., Moore R. K.,

and Fung A. K. (1982), “Microwave Remote Sensing: Active and

Passive. Volume II: Radar Remote Sensing and Surface Scattering

and Emission Theory”, Addison-Wesley Publishing Company.

The RSR is a novel technique that estimates surface properties from surface echo amplitude

statistics [4, 5, 6].

Total Surface Power

Reflectance (Pc)

Scattering (Pn)

2. LANDING RISK ASSESSMENT

Correlation coeff. (ρ) between

the empirical amplitude

distribution and the fit. It

estimates the agreement of the

real surface pattern with the

assumption of the statistical

model (at a horizontal scale of

~20 km for SHARAD).

ρ ↗ w/ surface homogeneity

ρ ↘ w/ surface heterogeneity

(roughness and composition)

Pc/Pn ratio illustrates the

degree of cohesiveness in the

signal received at the antenna.

A high Pc/Pn ratio does not

necessarily indicates a specular

surface. Pn fades 1/h2 faster

than Pc with the propagation

distance (~300 km in altitude)

Then, a high Pc/Pn associated

with a weak Pc might indicate

very rough surfaces.

Surface Properties Relationship Roughness Horizontal Scale 2014 Landing Ellipses

Pc = - 11.7 dB

Pn = - 11.8 dB

Pc/Pn = + 0.1 dB

ρ = 97.8 %

E05

E08 Pc = - 14.7 dB

Pn = - 10.6 dB

Pc/Pn = - 4.1 dB

ρ = 97.8 %

E09 Pc = - 15.7 dB

Pn = - 13.1 dB

Pc/Pn = - 2.6 dB

ρ = 97.4 %

E17 Pc = - 16.8 dB

Pn = - 10.9 dB

Pc/Pn = - 5.9 dB

ρ = 97.6 %

CONCLUSION

The surface echoes acquired all along a survey track are gathered in

successive packets (~1000 observations = ~20 km along-track for SHARAD).

Their amplitude distribution is best-fitted with a theoretical statistical

envelope [8] providing the coherent (reflectance) and incoherent (scattering)

signal components.

Reflectance (Pc) is mainly

sensitive to surface permittivity

and deterministic structures.

Coherent & Specular

+++ Permittivity

(density, composition)

+++ Thin deposits

+ Height distribution

Scattering (Pn) is mainly

sensitive to non-deterministic

structures.

Incoherent & Diffuse

+++ Height distribution

+++ Slope distribution

+++ Void in the near-surface

+ Permittivity

~20 km

Pc=r

2e−(2 kσ

h)2

R = Fresnel coefficient

k = Wave number

D = footprint diameter

H = altitude (~300 km)

ε = Surface permittivity

σh = Surface RMS height

l = Surface correlation length

Signal components (Pc and Pn)

can be related to surface

properties through backscattering

models. We usually obtain a 2

equations and 3 unknowns

system. For instance:

From the Small Perturbation

Method and the Physical Optic

[4, 8]:

From the Small Perturbation

Method only (restricted to

surfaces with σh < 0.75 m) [4, 8]:

The horizontal scale (HS) of the roughness

parameters are not well known. σh is

thought to be measured over several

wavelengths (λ = 20 m). l is closer to the

footprint radius (3000 m). Roughness

obtained from altimetry at various HS

shows some features (black arrow) invisible

at short scales are also not visible on Pn.

i.e. HS for SHARAD might also be closer to

several decameters.

Information content about

surface properties is important in the

radar signal. It can be constraint

with the RSR.

Relative risk assessment for

insight landing ellipses considering

surface permittivity and roughness.

Figure. Final 2014 InSight landing ellipses (orientation varies with launch

window) and SHARAD orbit tracks (black line). Terrain types in different colors

(green is smooth, orange is etched, red are steep crater walls or highland

scarp). Background is THEMIS thermal mosaic (From [2]).

We extracted surface amplitude histograms

for each of the 4 final landing ellipses (red).

Surface homogeneity (from ρ) is higher

than average. Pc/Pn ratios does not match

the application domain of the SPM. As a

preliminary assessment, we propose a

relative classification of the landing

hazard for each ellipse based on the

relative Pc and Pn. By increasing order:

E05: Less risk of soft material (ε > 3).

Intermediate roughness. Possibly covered

by two distinctive units.

E09: Low permittivity but the lowest risk of

rough surface.

E08: Intermediate permittivity but risk of

higher roughness.

E17: Risk of both low density (ε < 3) and

higher roughness.

NEXTToward an absolute/quantitative

risk assessment:

IEM and fractals backscattering

models to extend the RSR to a

wider set of terrains.

Characterization of the measured

roughness horizontal scale.

Constrain the signal calibration.