Skip to main content

Featured

Space Explorers Book 7: Answers and Explanations

Memunic Space Explorers › Book 7: Memoon and the Search for Another Home › Answers and Explanations Answers and Explanations Habitable means life could live there; inhabited means life does live there. A world can be habitable and empty, as an unoccupied flat has a roof and water but nobody in it. The Moon has almost no air, so water would boil or freeze, and its gravity is only about one-sixth of Earth’s, too weak to keep air. Accept also: no greenhouse blanket, so it swings between very hot and very cold. Venus. It matches Earth in size and roughly in mass, but its thick carbon dioxide air traps heat, making it hot and crushing. “Twin” is only true for size and weight. A dip of 1 per cent means (planet width divided by star width) squared is 0.01, so the ratio is 0.1: the planet is about a tenth as wide as the star. Repeats at the same gap show the planet is going round the star in 12 days, and rule out a one-time event such as a passing object. A planet shines only by reflected l...

Space Explorers Book 9: Chapter 9 – The Honest Answer

Memunic Space Explorers › Book 9: Memoon and the Clockwork Sky › Chapter 9: The Honest Answer

Chapter 9: The Honest Answer

Illustration for Space Explorers Book 9: Chapter 9 – The Honest Answer

In this chapter: tell random from systematic error, report a result with its uncertainty, and separate what was measured from what was inferred.

At 5:54 pm the next evening, six pairs of eyes were fixed on the north-west.

The sky was a clear, pale orange fading to blue. Memoon had her notebook open to the page with the big prediction. Kachui had set his phone to display the exact time, synchronised that morning with the official time signal. Beautine had her wristwatch, a gift from her grandmother, in which she had great faith. Noodles had a stopwatch he did not know how to use.

“Five fifty-four,” Memoon said. Nothing happened.

“Five fifty-five,” said Beautine, looking at her wrist. “Nothing.”

At 5:55 by Beautine’s watch, a light appeared over the north-west horizon, bright and steady.

“There,” cried Beautine. “One minute late! Only one minute!”

Kachui, who was watching his phone, said nothing for a moment. “By my clock it is 5:57 and ten seconds.”

“My watch says 5:55,” said Beautine.

“Then one of us is wrong,” said Kachui, “by two minutes.”

The station climbed through the pale sky and faded in Earth’s shadow, as it had the night before. When it was gone, the club stood around the notebook.

“The prediction was 5:54,” said Memoon. “By Kachui’s clock the station appeared at 5:57. That is a miss of three minutes.”

“By mine it is one,” said Beautine.

“Whose clock is right?” said Noodles.

Kachui took out a small slip. “This morning I set my phone against the national time signal. I will check the watch.” He checked it, and was gentle in telling her. “Beautine, your watch is slow by about two minutes.”

Beautine clutched it. “My grandmother’s watch.”

“It is a lovely watch,” said Kachui. “But every time you read it, it is slow by the same amount. That is a different kind of mistake from a random slip of the hand. We call a mistake that pushes every reading the same way a systematic error. A random error is when readings scatter up and down by chance, like my pencil mark on a protractor.”

“So my error is systematic,” said Beautine, with interest, “and Memoon’s three minutes?”

“Let us find out,” said Memoon. “This is a good one. I made a prediction and I missed. I do not want to call it clockwork, and I do not want to call it a failure. I want to know why.”

She made a list in the notebook, and as she wrote, she thought aloud.

“One. The period. I used 92.9 minutes. But 92.9 is itself rounded. If the real value is 92.85 or 92.95, then fifteen orbits could be 0.75 minutes more or less. That would be an error of about three-quarters of a minute, in either direction.”

“If I had used the rounded 93,” said Noodles, working it out on his paws, “it would be 0.1 more each orbit, fifteen orbits, one and a half minutes.”

“Right. A small rounding becomes larger when you multiply it by many orbits. That is a good lesson in itself.”

“Two. The orbit is not exactly the same every day. Thin air at that height drags on the station and slowly lowers it. Every few weeks its engines lift it again. After one of those boosts, the timing changes. I did not know that when I made my prediction.”

“Three,” said Kachui, “you compared the time it first appeared with the time it was first predicted. But the pass is longer than that. If a hill or a building hides the first part of the sky, the station appears a little late. The first time we saw it was not the time it came over the horizon.”

“I did not think about that either,” said Memoon, and wrote it down.

“So you have three sources of error,” said Simbha, “and a total miss of three minutes. We do not know how it divides between them.”

“No. I cannot say which part was which,” said Memoon. “And that is fine. The important thing is that I can say how big the miss was and what might cause it. I cannot say it was exactly right.”

Kachui nodded. “This is why you should always state a result together with a range. Not ’the station will appear at 5:54’, but ’at 5:54, give or take three minutes’.”

He turned to a fresh page. “The Jupiter log needs the same treatment.”

They went through the table. Over the nine nights, six observations of Ganymede’s greatest distance east or west had been recorded: Noodles had written fourteen and fifteen and fourteen and a half and fifteen and a half, Kachui fifteen, Memoon fourteen and a half. There was also an odd one: one reading of seventeen, from the night Noodles knocked the tripod.

“The knock,” said Noodles, ashamed. “I did not mean to.”

“We do not hide the seventeen,” said Memoon. “We mark it as an outlier, and say why. A reading that is far from the others is not thrown away secretly. If we remove it, we say so and give the reason. Here the reason is clear.”

She averaged the other six readings. “Eighty-eight and a half divided by six. Fourteen point seven five. The lowest was fourteen and the highest fifteen and a half, a spread of one and a half, so half of that, 0.75. So we say 14.75, plus or minus 0.75 steps.”

“And the real value?” said Beautine.

Nimo drifted up. “Astronomers give Ganymede’s orbit as about 1,070,000 kilometres from Jupiter’s centre. Jupiter’s radius is about 71,500 kilometres. That is a little less than fifteen radii: 14.97, rounded.”

“Inside our range,” said Kachui. “We should check that the number agrees with ours within its uncertainty, which it does. That does not prove we are right. It tells us we have not found a disagreement.”

“Is there a systematic error in our log?” said Memoon.

Kachui looked at the notebook for a long time. “Maybe. Ganymede at its greatest distance east is usually fifteen in our log. On the west it is fourteen, a little lower. If Jupiter’s centre on my scale was a little off, every reading east would be too big and every reading west too small. The centre mark on my strip might be half a step off.”

He flipped the plastic strip over, aligned it with Jupiter again, and measured the moons one more time. East peaks were fourteen and a half. West peaks were fourteen and a half.

“I think,” he said, “we found a systematic error. The zero of our scale was half a step out. It did not change the period at all. But it shifted every reading the same way. Now both sides agree.”

Two kinds of error

Random error Systematic error

Scatter around the true value. Every reading is shifted the same way.

Averaging helps. Averaging does not help; check the instrument. Green dot = true value; orange dots = readings.

Random error scatters readings; systematic error shifts them all.

“So the period was safe,” said Memoon, “but the amplitude was not.”

“Which means,” said Kachui, “that we should think about which errors matter for which result. A slow watch does not change the period of Jupiter’s moons if the same watch is used all the time. It changes the clock time of any event.”

Nimo bobbed. “May I tell you about a man who used a clock error as evidence?”

They gathered round. “In the 1670s, a Danish astronomer named Ole Rømer was working in Paris. He was timing the moment Io disappeared into Jupiter’s shadow, an eclipse that happens every 1.77 days. He expected a steady pattern, like a clock. But when Earth was moving away from Jupiter, the eclipses came later than the timetable, and when Earth was moving closer, earlier. Over months, the difference grew until it was about eleven minutes ahead or behind a steady clock, twenty-two minutes from one extreme to the other.”

“A systematic error?” said Beautine.

“A systematic effect,” said Nimo. “It did not scatter. It followed Earth’s position. Rømer concluded that light takes time to travel. When Earth is farther from Jupiter, the light showing the eclipse has farther to go and arrives later. He estimated that light would take about twenty-two minutes to cross Earth’s orbit. Today we measure about sixteen and a half. His number was off by a third, but the idea was right.”

“What did he measure?” said Kachui.

“The times of eclipses, with a clock.”

“And what did he infer?” said Memoon.

“That light travels at a finite speed. He never saw light moving. He worked it out from what he measured. That is the difference between what is measured and what is inferred. A measurement is a reading from an instrument. An inference is a conclusion that needs reasoning, assumptions, and sometimes other evidence. An inference can be wrong even when the measurements are good. So it should be tested another way.”

“Was it?” said Beautine.

“Later, by other astronomers, yes.”

Memoon stared at the notebook. “Then I should apply it to our Night 15 prediction for Ganymede. Our curve says its position should be about minus five steps on Night 15, west of Jupiter. But the period is 7.2 plus or minus 0.3 days. Over two periods, that error grows. If the period were 7.5, the position would be very different.”

She worked it out. “Plus or minus seven steps. The prediction is minus five, give or take seven. So the real answer could be anywhere between plus two and minus twelve.”

“That’s almost useless,” said Beautine.

“It is honest,” said Kachui. “And it tells us what to do. The longer we watch, the more cycles we have, and the more precisely we know the period. If we know the period to a tenth of a day instead of three tenths, the range shrinks to a third. Errors in prediction grow the further you extrapolate.”

“Like the station,” said Noodles. “If I want to know when it will pass, I should use observations from just before, not yesterday.”

“Yes,” said Memoon. She smiled. “That is what we will do on Saturday.”

Beautine looked at her watch. “Perhaps I should get it fixed.”

“Perhaps,” said Kachui, “you should keep it. It is a lovely watch. You should just write on it how slow it is.”

Science Corner A random error scatters readings up and down by chance; repeating and averaging reduces its effect. A systematic error shifts every reading the same way; averaging does not remove it, but checking the instrument against a standard can reveal it. A result should be stated with its uncertainty, for example 14.75 ± 0.75 steps. The mean is the sum of readings divided by their number, and the spread (the half-range here) shows how much they scatter. An outlier is a reading far from the rest. It should be kept in the record, flagged, and removed from an average only with a stated reason. A quantity that is measured is read from an instrument; one that is inferred is concluded from measurements through reasoning. Rømer measured eclipse times and inferred that light takes time to travel; today’s value for light crossing Earth’s orbit is about 16.6 minutes. Errors in a prediction grow as it is extended further from the data. A small error in the period, multiplied by many cycles, becomes a large error in position or time. Story note: Rømer’s historical estimate (about 22 minutes for light to cross Earth’s orbit) differed from the modern value of about 16.6 minutes. The “eleven minutes early or late” figure follows the usual retelling of his results. The club’s numbers (14.75, ±0.75, ±7 steps) are worked from the log in the story; the pass times are invented. Where this idea stops working: a range such as ±0.75 describes the spread of readings, not a guarantee that the true value lies inside it.

Space Detective 1. A clock runs 2 minutes slow all the time. Does this change a measurement of the time between two events? Does it change the measured clock time of one event? Explain. 2. Four readings of an amplitude are 14, 15, 14.5 and 15.5 steps. Find the mean and the half-range, and write the result with its uncertainty. 3. Draw a graph with a smooth curve through five scattered points, and mark one point you would flag as an outlier, with a note explaining your decision.

Try it: Time It Twice Time: 25 minutes.

Adult: An adult should supervise, and nothing breakable should be on the table. You need: a stopwatch (or phone), a ball, a table edge, a helper.

Do: Roll a ball off a table and time how long it takes to hit the floor, ten times. Write all readings. Find the mean and the spread. Then check your stopwatch against an online time signal.

What to expect: readings scatter by about 0.1 s because of reaction time. If the clock is off, all readings shift together. Cloudy or low-resource day: this is an indoor activity. If no stopwatch is available, count “one-Mississippi” for slower events, like a rolling toy car, and compare with a friend’s count.

Limitation: counting is crude and the spread will be larger.

Next time: Data Night is on Saturday. Can the Sky Club make a prediction honest enough to be useful?

← Chapter 8: The Station’s Timetable · Book 9 contents · Chapter 10: The Clockwork Sky →

Keep exploring with Memunic Space Explorers

Continue with Book 9: Memoon and the Clockwork Sky, or see all twelve books in the series. Find more Memunic ebooks on MemuniCity.

Comments

Popular posts from this blog

CA-7 Job Status and Failure Reason Codes

Here are the various status codes that You may see for jobs in the Queue or Logs in CA-7. R-C0002 R-C0003 R-C0099 R-C0255 R-JCLER A-U0002 A-U0004 A-U0015 A-U0043 A-U4095 A-SOC7 A-U4000 Reason Code/Job Status Code explanation: The indication of the job's current disposition. Values and their meanings are: A-Snnnn-- The job has failed with a system abend. A-Unnnn --The job has failed with a user abend. C-Cnnnn --The job has completed successfully and is going through completion processing. E-ARFATT-- An error occurred during the attempt to attach ARF definitions. This can result if the ARFSET specified is not defined. A diagnostic message may have been written to the CA-7 log, the OS console, or to the trailer queue for the job (in which case it can be displayed using LQ,JOB=x,LIST=ALL). E-ARFERR ---The job encountered an error in the ARF processing. Do a LARFQ command for the job to get more specific information. E-SEGERR --The job has been requeued...

Whistling Winds and Waterfalls: 4 Days of Magic in Shillong

Nestled in the clouds, where the scent of pine trees mingles with the rhythm of waterfalls and the tunes of local guitar-strumming youth—Shillong is more than a place. It’s a vibe. If you’ve got 3 nights and 4 days, you’ve got just enough time to fall in love. Here’s how to make the most of it—food, fun, fog and fantastic views included. Day 1: Shillong, We’re Coming for You! You land in Guwahati, barely out of the airport when the hills start calling. You hop into your cab, Spotify on, windows down, and the three-hour ride to Shillong begins. Just when you think it can’t get better, you stop at Umiam Lake —a lake so calm, it looks like someone ironed the water. By evening, you’re sipping hot chai in Laitumkhrah, maybe nibbling on spicy momos or pork jadoh from a roadside stall. You’re checked into your charming hotel ( Tripura Castle if you fancy old-world royalty, Ri Kynjai if you’re into lakeside poetry). The breeze whispers secrets of waterfalls and pine forests. Sleep well, drea...

Installing and Configuring Hadoop on Ubuntu as a Pseudo Distributed Single Node Cluster

Briefly what is psuedo-distributed single node Hadoop cluster: It is a Hadoop distributed cluster having Single Node. All the five daemons (NN,DN,JT,TT,SNN) are running as individual JVM (java process) instances. We will get all distributed features of Hadoop. This mode is used to develop, test, and staging (before production) Map Reduce programs quickly. It writes all the system, daemon and user logs to different files on the cluster To install Hadoop in this mode, we have to modify the hadoop-env.sh, core-site.xml, hdfs-site.xml, mapred-site.xml. hadoop-env.sh contains all the Environment variables which are required for running Hadoop. All the default values are fine, Only we have to modify the JAVA_HOME=“The location of Java Home Directory” Core-site.xml contains all the cluster specific information like io, temp directory, sorting, file system uri etc. Hdfs-site.xml contains all the parameters which are specific to HDFS like name node directory, data node directory...