Human Spaceflight: What Value to Science? (Pt. 2)

Robots vs. astronauts

John Young at Plum Crater, Apollo 16, April 1972. (NASA)

The discussion at Space Politics got me thinking about the scientific value of human spaceflight.  Although there are many reasons for humans to go into space, I also believe that humans bring unique and non-duplicative skills to scientific exploration as well.

In part one of this post, I discussed how the capabilities and experience of the Apollo missions led to a revolution in our understanding of the history of both Moon and Earth and gave new insight into the process of evolution and in turn, our origin.  In this post, I want to look ahead to the value humans bring to future exploration, particularly to the scientific exploration of the Moon and their critical role in lunar return, resource development and ultimately, settlement.

A common article of faith in many academic and space circles is that robotic spaceflight is the preferred method of scientific exploration.  Many famous space scientists (including James Van Allen and Carl Sagan) preached the superiority of unmanned missions to human ones.  Indeed, many phenomena in space (such as plasmas and magnetic fields) cannot be sensed directly by humans or in some cases (e.g., detecting the tenuous lunar “atmosphere”), the presence of people interfere with the property being measured.  I agree that some scientific activities cannot or should not be done by people.  But in other areas of science, human presence is not merely beneficial, it’s critical.

The Moon is a natural laboratory where we will answer critical scientific questions.  The “field” is the world in its natural state, where the phenomena we study are on display and where we learn the key facts that permit us to reconstruct past processes and histories.  Field work is not merely a matter of picking up rocks or taking pictures.  It is the conceptual visualization of the four-dimensional (three spatial dimensions plus time) make-up of planetary crusts.

A good example of the differences in capabilities between humans and robots is illustrated by the experience with the Mars Exploration Rovers (2003-present).  These  machines have traversed many kilometers of terrain, examined and analyzed rock and soil samples, and mapped the local surface over the course of five years.  Many gigabytes of data have been returned by these rovers, giving us an unprecedented view of the martian surface and its geology.  They are truly marvels of modern engineering.

Yet after all this exploration, we are unable to draw a simple geological cross section through either of the two MER landing sites.  We do not know the origin of the bedded sediments strikingly shown in the surface panoramas, whether they are of water-lain sedimentary, impact, or igneous origins.  We don’t know the mineral composition of rocks for which we have chemical analyses; such information is crucial to determine processes and origins.

After five years of Mars surface exploration, we do not know things about the field site that a human geologist could determine after an afternoon’s reconnaissance.  In contrast, we have an incredibly detailed conceptual model, albeit incomplete, of the geology and structure of each of the Apollo landing sites.  The longest stay on the Moon for these missions was three days, most of which was spent inside the Lunar Module.

A robotic rover can be designed to collect a sample, but it cannot be designed to collect the correct sample.  Field work involves posing and answering conceptual questions in real time, when emerging models and ideas can be tested in the field.  It is a complex and iterative process; we sometimes spend years at certain field sites on the Earth, asking and answering different and ever more detailed scientific questions.  Our objective in the geological exploration of the Moon is knowledge and understanding.  A rock is just a rock, a piece of data.  It is not knowledge.  Robots collect data, not knowledge.

It has been argued that planetary exploration robots are controlled by people so human intelligence guides the robot explorer.  Having done both types of exploration in the field on Earth, I contend that remote teleoperated robotic exploration is no substitute for being there.  All robotic systems have critical sensory limitations – important sensory aspects as resolution, depth of field, and peripheral vision.  They have even greater limitations in physical manipulation, an extremely important aspect of field work, where picking a sample, removing some secondary over coating and examining a fresh surface is an important aspect of work in the field.  The makers of the MER rovers recognized this need by including the RAT (Rock Abrasion Tool) to create fresh surfaces; it became worn down and unusable after a short period of operation.

Ultimately, we need both people and machines to explore the Moon and other planets.  Each has their appropriate skill base and limits.  Machines can gather early reconnaissance data, make preliminary measurements, and do repetitive or exhaustive manual work.  But only people can think.  And thinking – and acting and working based upon the results of that thinking – is what field work is all about.
About Paul D. Spudis
Paul D. Spudis

Paul D. Spudis is a senior staff scientist at the Lunar and Planetary Institute in Houston, Texas. His website can be found at The opinions expressed here are his own and do not reflect the views of the Smithsonian Institution or his employer.

Read more from this author

Comment on this Story

comments powered by Disqus