PREDICTING THE FUTURE
It may seem hard enough trying to reconstruct what has happened to the environment in the past, but at least there are some concrete indications and signs available to help. Predicting the future is even more difficult, because no one knows for certain what will happen, and there can always be surprises. Some things like the changing seasons can be predicted with reasonable certainty, but even seasonal weather patterns may change from year to year. Other events such as the occurrence of an earthquake or the passage of a cyclone (hurricane) over a particular place may seem totally unpredictable in the present state of scientific knowledge in the region.
Those events, like the seasons, that are cyclical and occur at regular
intervals, make it easier to predict some kinds of effects. Some cyclical
events are the result of astronomical or celestial processes, such as:
- day and night, the rising and setting of the sun;
- the phases of the moon;
- the tides;
- seasonal changes over the year.
There are also many biological events associated in one way or another
with these physical cycles. Some examples are:
- the flowering and fruiting of many plants;
- the breeding of some animals;
- bird migrations;
- some movements of fish around the reef.
Other events are random, that is they may occur at any time, and each
occurrence has no relation to the preceding one. An example of a random
event is flipping a coin to see if it comes up heads or tails (one side
or the other); which side comes up is a random event. Such events could
as easily occur twice in a row as not to occur at all for a long period.
However random events can differ in the frequency of their occurrence.
You can flip a coin many times or only a few times without having any effect
on the random appearance of one side or the other. Some examples of random
- the accidental arrival of a new species on an island;
- the path of a cyclone;
- the place where lightning strikes.
While the future can never be predicted with certainty, there are ways to have an idea of what is most likely to happen. One of these is called extrapolation, which means predicting what will happen in the future by projecting the trend of what has happened in the past and what is happening now. It is necessary to assume that the conditions that have occurred up to now will not change, or, it they do, that they will change in some known way that can be adjusted for. For instance, if a thousand hectares of forest have been cleared on an island every year for the past five years, you might extrapolate that five thousand more will be cut in the next five years. If each year the amount of forest cut increased, say by 500 hectares a year, you might expect the amount cut to go on increasing, assuming no change in conditions.
Known Predicted Year 1980 1985 1990 1995 2000 2005 2010 2015 Steady Cut 1000 1000 1000 1000 1000 1000 1000 1000 Increasing Cut 1000 1500 2000 2500 3000 3500 4000 4500
Most extrapolations have some limit beyond which they will not hold. If all the remaining forest will be cleared in three years, then cutting must stop at that limit if not before.
Population projections for some future time are usually based on an extrapolation from the past and present rates of population growth. They make it possible to plan for the number of schools to build, the number of jobs or houses that will be needed, and so on. However, if there is an unexpected change in the rate of population growth, then the extrapolated population figures will need to be revised.
Sometimes an extrapolation will show that a present trend must be changed. If 1,000 hectares of new agricultural land are being cleared every year, and there are only 10,000 hectares of good unused land left on the island, then by the end of ten years some alternative will have to be found to clearing more land, such as making better use of the land already cleared.
Extrapolations must be used with care, since sometimes the present trend itself will bring about changes in conditions, making the extrapolation false. In some places fishermen have been increasing their catch every year by adopting new fishing technologies, and an extrapolation from present trends would suggest that more investment in fishing equipment was worthwhile. But the new technologies may themselves lead to overfishing, producing a fall in the catch rather than an increase.
Many development projects have failed because they were based on overly optimistic extrapolations of the size of the resource or the production capability without a careful study of limiting factors.
Another measure useful in predicting the future is probability, which is the chance that something may happen in a certain period of time or during a certain number of events. In each flip of a coin, the probability of each side landing face upward is 50% (percent), and for each flip the probability is the same. The fact that the previous flip or even the previous 5 flips came up heads will have no effect on the probability of heads on the following flip. If records show that there has been a cyclone on the average approximately every 10 years, than the probability of a cyclone in any given year is 10%. Each year there is one chance in ten that there will be a cyclone, but over 10 years, the chance of having at least one cyclone approaches 100%.
Probabilities are used by insurance companies to calculate the cost of insurance. Based on past statistics say of automobile accidents, they calculate the probability of any one car having an accident to determine how much to charge each driver for insurance. If the company's calculations are correct, what it pays out for the few accidents that happen will be more than made up by what it receives from all the insured drivers.
While a probability estimate does not predict the future, it can give a measure of the possibility of something good or bad happening in a given period of time, and this makes it possible to estimate the chances of success or the risk of failure. It is then possible to include some provision in a project to cover these risks, just as an insurance company makes provision for its losses.
For example, suppose you are developing agricultural land in a river valley where the river is known to flood approximately every ten years. You thus know that there is a 10% probability each year that your plantings will be damaged by flooding, but you may feel it is worth the risk if the yield for the other 9 years will more than make up for the one year lost. However, this does not mean that you will have 10 years before the next flood. There is even a slight chance that there could be floods 2 years in a row, perhaps followed by a longer period without flooding. You might not build a house where there was a probability of flooding every 10 years, but you might in a place where the probability was only once in a hundred years.
Some probabilities can be calculated on the basis of what causes the event. In flipping coins, it is clear that there is an equal chance of each side landing upward. Others can be predicted on the basis of historical records, in which a probability is extrapolated from the frequency of past occurrences of an event. If records show that over the last 70 years and island was hit by 7 cyclones, then the annual probability of a cyclone can be calculated as 10%.
Using predictions of the future
No system for predicting the future will be right in every case or all of the time. The kinds of methods described here can increase the chances of being right and reduce the possibilities of a major error. They are therefore a useful part of the planning process.
It is important in using extrapolations or probabilities to evaluate not only the chances of success or failure, of good conditions or natural disasters, but also the consequences of being right or wrong. It is normal to accept a greater risk where the consequences are marginal than in situations that may be matters of life or death.
For instance, if you already have several fields under cultivation, you might take the risk of planting an area subject to flooding, because even if there is a flood, only a small part of your harvest would be lost. However, if you only could afford to plant one field, you would probably choose a safer place to avoid the risk of losing everything.
Similarly, a risk that would be acceptable for a crop or other agricultural development might not be acceptable for houses or school buildings. A 20% chance of damage might be acceptable for an annual crop, but not for a perennial crop like coffee. You would do everything possible to make a safe boat to transport your family, but probably less for a raft to transport building materials.
A relatively small probability of a major disaster may weigh more heavily in planning than a much greater risk of minor damage. A serious earthquake may occur only once in a hundred years or more, but failure to plan for it could lead to great loss of life.
Once a wise use of extrapolations and probabilities has been made using
the best environmental information available, the challenge for environmental
planning is to bring together this information with the best estimates
of the consequences of various events and courses of action to arrive at
plans that are reasonable, productive and safe.
Why is it useful to try to predict the future of the environment?
Is predicting the future based on magic or science?
What are some things that can be predicted easily?
What are some things that are very hard or impossible to predict?
What is meant by extrapolation into the future?
Can you give some examples of extrapolation?
What does a probability measure? Give an example.
How can probabilities be used in planning?
What is the annual probability of a cyclone, hurricane or typhoon in your area?
What kind of natural disaster has the greatest probability where you live?
What preparations do people make for such future events?
Can you predict what your town or village will be like in 10 years time, using extrapolations and probabilities?
What do you think your own family will be like in 10 years? How many adults and how many children, and of what ages?