Chapter 120: Important But Non-urgent Weather Forecasting
“Really? Is that true?”
Two hours later, Ye Duzheng, who had gone to the Chinese Academy of Sciences Computing Center that had just been officially established to retrieve the calculation results, couldn’t help but open the calculation results to check them on the road.
Regardless of the specific content, at least what was calculated was something like that.
It calculated the position, intensity, and movement path of the future weather system.
When he returned to the Meteorological Bureau and handed the results calculated by the Computing Center to Tu Changwang, Tu Changwang had a similar thought to his: “It seems impossible to calculate it in two hours.”
But they did calculate it.
Just right, for the next two weeks, we will take the daily numerical weather predictive model and parameters to the Computing Center every week and let them calculate it.
Then compare their calculation results with the observation results from local meteorological observation stations.
See how accurate their calculations are; if the accuracy is higher than our current calculation method, we can completely use their calculation method.”
After hearing this, Ye Duzheng said: “Director, there is still something a bit off.
Don’t you think the results provided by the Computing Center are too precise?
Even if they use computers to provide us with calculations, even the Soviet Union’s BESM-2 couldn’t achieve this precision.”
Tu Changwang said: “We don’t need to worry about that.
We don’t care how the Computing Center achieved it; we only need to care whether what it calculates is accurate.”
He went back and pondered Dean Guo’s speech again; Tu Changwang guessed whether China had achieved a technological breakthrough in some aspect, hence the extreme caution.
Furthermore, thinking about how from Hua Luogeng to the entire mathematics institute of the Chinese Academy of Sciences, a large number of mathematicians had disappeared, said to have been transferred to Yangcheng, but who knows where they were actually transferred.
These mathematicians disappeared in mid-1960, and by early 1962, the Computing Center was officially established, right?
Ye Duzheng said the Computing Center only has two people; how do you know it’s really only two people? Tu Changwang thought, perhaps through some pipe, you give them input, they put it into the pipe, and the pipe delivers what you need to calculate to other rooms.
The two-person Computing Center is just what’s shown openly to you.
Including the missile technology that the Soviet Union “forcibly” purchased from China last year, Tu Changwang had heard about it, estimating that it was also a result of China’s technological breakthrough.
Regarding these, that’s why he went to find Dean Qian earlier, trying to verify his guess, and sure enough, it was similar to what he guessed; Dean Qian had also been transferred away.
Therefore, Tu Changwang guessed that China had achieved an unprecedented breakthrough in the field of numerical analysis, and moreover, because the breakthrough was extraordinary, there was an exceptionally strict secrecy requirement.
The Computing Center might be mathematicians led by Hua Luogeng using human brains to provide calculations for them, he guessed.
Therefore, since there was a guess, there was no need to dig deeper.
Tu Changwang himself did not want to dwell on issues of details anymore, nor did he let his subordinates inquire.
Two weeks later, Tu Changwang organized a meeting with the experts from the Meteorological Bureau.
“Come on, Duzheng, you do the summary.”
Ye Duzheng walked to the blackboard, which was covered with papers; the papers showed comparisons of their calculation results, Computing Center calculation results, and actual observation results data.
“This is the 24-hour weather prediction results over the past two weeks. The first column is the summarized results reported from local meteorological observation stations after observation, the second column is the results calculated using our previous calculation method, and the third column is the results calculated using the latest numerical weather predictive model.
Among them, the accuracy of the second column is 65.2%, and the accuracy of the third column is 90.3%, far higher than the second column.”
After speaking, Ye Duzheng walked to the second blackboard:
“Here, starting from last week, we extended the weather prediction time to 48 hours.
Same as the 24-hour one next to it: the first column is the observed actual results, the second column is the previous predictions, and the third column is the predictions using the new method.
This time the gap between the two sides is even larger; the accuracy of the second column has fallen below 50%, while the third column still maintains about 70% accuracy.
So my summary is that we can use the new method to replace the old method to provide weather forecasts for the people nationwide and local governments.
On the other hand, we need to fully excavate the potential of the Computing Center.
In my view, our current predictions are not done enough.
Our numerical weather predictive model has just started, and our meteorological observation network is still very sparse, mainly concentrated in eastern cities, with almost no high-altitude and ocean data.”
The experts present who knew that the Meteorological Bureau was cooperating with the Chinese Academy of Sciences Computing Center, even though they knew in advance, were still shocked after seeing the results and hearing such a huge contrast in the data.
Those who didn’t know were even more shocked inwardly.
Among them, the most obvious was Zhu Kezhen.
He was the leading figure in China’s meteorological field, a true master-level character.
At this time, he was working at the Institute of Geography of the Chinese Academy of Sciences; of course he knew about the establishment of the Computing Center, but he didn’t expect it to have such power.
Because his current work mainly involved crop observations, hosting the establishment of a nationwide phenological observation network at the Institute of Geography of the Chinese Academy of Sciences, he had already moved away from frontline meteorological work.
The work he was currently responsible for basically did not require using the resources of the Computing Center.
Plus, from the establishment to this meeting today, it was only two weeks, so Zhu Kezhen also found it hard to hear about the extraordinary aspects of the Computing Center from his colleagues.
But this time, he had witnessed it with his own eyes.
“I have a suggestion: you must try your best to excavate the potential of the Computing Center.
Back when we were doing the numerical weather predictive model, we were aiming to track foreign advanced achievements and frontline applications abroad.
Thinking that China would sooner or later have its own computers, and we could use the numerical weather predictive model.
But because it was in the 1950s, this work was done on the premise of beautiful expectations for the future.
In reality, we weren’t clear on exactly when we would have our own computers, or when they would be allocated to the Meteorological Bureau for weather prediction work.
So everyone did this work with somewhat pre-research nature, without delving the entire model as deeply as possible.
Just simple filter equations and single-layer quasi-geostrophic model, without considering vertical motion and non-geostrophic effects.
Not enough consideration in terms of precision and applicability.
Of course, I’m not pursuing everyone’s issues; after all, these are reality factors.
It’s like asking me now to prepare for something that might happen who knows how many years later; I would do it, but I wouldn’t put particular effort into it.
Now that we have it, our models also need optimization and adjustment to better fit the actual situation.
Additionally, regarding exactly what precision the Computing Center can calculate to, I think you need to focus on observing that.”
Obviously, although Zhu Kezhen had never managed the Meteorological Bureau’s work, when he spoke, everyone listened very seriously.
His status in the meteorological field was there.
Not to mention that what he said hit the nail on the head.
He immediately pointed out where their problems lay.
At this time, Tu Changwang added: “Academician Zhu is absolutely right; this is also one of the most important subsequent tasks for our Meteorological Bureau.
We will definitely optimize the existing models.
Last year we obtained a batch of America Meteorological Society published academic journals, which contain many papers related to numerical weather prediction, and we are organizing people to research and extract the core content.”
At this time point, JAM had not yet been founded; mainly JAS and MWR.
The former is basic research in atmospheric science, while the latter mainly covers weather analysis, prediction technology, and business applications.
The latter has a lot of highly academic numerical forecasting research content.
“According to our recent research, America is already using multi-layer quasi-geostrophic models, at least three-layer structure, including atmospheric vertical structure and large-scale circulation.
Their resolution has reached the 200-kilometer level, capable of accurate predictions for three to five days in the future.
Indeed, as Academician Zhu said, we need to optimize our models.
We need to consider more factors.
However, regarding the meteorological observation network, we are not too sure what level of computation the Computing Center can provide.”
Zhu Kezhen knew a bit more than Tu Changwang, but he only knew that China had achieved certain breakthroughs in the transistor computer field; he didn’t know the specific situation.
He said: “If you don’t know, go ask; find Dean Guo, he definitely knows the situation.”
Zhu Kezhen continued: “Additionally, regarding specific model optimization, I have some ideas you can refer to.
First, introduce as much data as possible; include all data observed by current meteorological observation stations into the model, like pressure, temperature, humidity, wind speed, etc.
For example, the quasi-geostrophic vorticity equation can divide the country into grids, then input pressure fields and temperature fields, and calibrate your model based on actual measurement data in the regions.
Second, optimize the model; I don’t expect us to jump to five layers at once, but at least three layers. Run a three-layer primitive equation model; local observation stations need to take on more tasks, like figuring out ways to measure high-altitude conditions.
This way, accurate predictions can be made for more complex situations.
Third, increase the prediction days: from one day to two days, two to three, even to five days.
Finally, our current meteorological observation network is still concentrated in eastern cities; we also need to pay attention to places like Southwest, Northwest, South China, North China. The Fifth Academy is researching and developing satellites; we need to make requests to them, so that when their satellites are launched, they help us capture high-altitude data.”
After the meeting, Tu Changwang ran to the Chinese Academy of Sciences to find Dean Guo.
Before he could speak, Dean Guo guessed his intention: “Old Tu, I know what you mean; don’t worry, your Meteorological Bureau will definitely have the highest priority at the Computing Center.
Additionally, have you replaced the weather forecast data you send to various places with the calculation results from the Computing Center?”
