The “Precision” in Precision Forestry
Zack Parisa
Zack Parisa
20 October, 2018 min read

This article was first published in the October 2018 issue of The Forestry Source.

This summer, the McKinsey consulting firm published a report entitled “Precision forestry: A revolution in the woods.” We eagerly read the report when it came out and several of our readers have asked us to provide a framework for thinking about “precision forestry.” After reading this installment of “Biometrics Bits,” you’ll have answers ready when your clients ask, “So what does precision forestry mean for us?” or “What is our technology strategy?”

The phrase “precision forestry” conjures up hazy visions of massive computers, swarms of drones, and intelligent automated harvesters. This flood of data is supposed to make us smarter and more efficient – but how can you cut through the confusion to choose the tools that fit your forest? With potential double-digit gains, it’s incumbent upon us as forestry professionals to consider integrating these new tools and practices. There are all sorts of promising new technologies and we need a framework for evaluating whether the gains outweigh the costs.

One of our readers (who asked to remain unnamed) suggested a vivid way to illustrate this dynamic and we thought it was worth sharing. In this article, we’ll play a game to develop an intuition about this tradeoff between increased precision and increased cost. With this perspective, you’ll be in a better position to evaluate the new generation of “precision forestry” technologies in the context of your unique forest.

Feeling lucky?

Let’s play a game. I’m thinking of a number between 1 and 100.  If you guess my number, you win $100. It costs $5 to play. Any takers?

Intuitively, this should seem like a bad deal. You only have a 1 in 100 shot of being right and $5 seems a bit steep. In fact, you can prove to yourself that this is a bad deal with some simple math. You can calculate an “expected value” which is the probability of winning (1/100) multiplied by the value of winning ($100). In this case, your expected value is $1. But when you subtract out the cost of playing ($5), you’re left with -$4. Down 4 bucks?! Step away from the game!

Let’s make this more interesting. The game still costs $5 to play, but you also can pay me extra to narrow down the range of possible numbers. Here are my prices:

So at the most extreme end, I’ll tell you the exact number I’m thinking of for the low, low price of $150. Any takers?

Of course not! It doesn’t make sense to pay more for information than the maximum potential payoff. Paying $150 (plus $5 to play) for $100 is a quick way to go bankrupt.

At the other end of the spectrum, for $3, I’ll tell you whether my number is in the range 1-20, 21-40, 41-60, 61-80, or 81-100. Is this worth it? Seems like we’re significantly improving our chances, but is that improvement worth more than $3? Our old friend, “expected value” comes to the rescue. Our probability of winning is now 1/20, so our expected value is $100 * 1/20 = $5.  But when we subtract out the cost of playing ($5) and the cost of the information ($3), we’re left with -$3. Not such a good deal after all. Let’s run this analysis for all of our options:

Reading off the “Net Expected Value” column (our expected value minus the cost of playing and the cost of information), we see that paying for the 2-number precision option is the way to go – we would expect to end up with $10.

Back to the woods

In many respects, this simple game is similar to the questions that you face when you plan an inventory or develop a management plan. The $5 represents your overhead (taxes, cost of capital, etc.). Just like in this game, precision exists along a spectrum and the more you pay (for more plots, etc.), the higher precision you can achieve. With higher precision information, you have the potential to improve your forest management outcomes. But you always have to keep an eye on the cost of obtaining higher precision (and those better outcomes) to make sure you’re optimizing for the net expected value.

One of the big takeaways from this game is that relying on rules-of-thumb could be costing you and your clients money. Without a framework to systematically decide which option is best, a guess one way or the other could really cost you. As the game demonstrates, there is a right answer for the level of precision you should be targeting and any other choice is taking money out of your and your client’s pockets. The correct level of precision will depend on the value of the decisions you’re making (the maximum possible NPV of your forest) and the cost of achieving various levels of precision.

So where does “precision forestry” come into all this? As the name suggests, the role of all this new technology is to enable you to achieve greater precision at a lower cost than traditional methods. This in turn enables better management and better outcomes for forest owners. Now that you have a framework for assessing the value of increased precision, you have a head start on making informed, quantitative decisions about your forest technology strategy.

One final aspect of the “precision forestry” conversation that is frequently ignored is the importance of tracking precision throughout the entire value chain. In our past “Biometrics Bits” articles, we’ve focused on precision in forest inventory, but growth models, volume equations, harvest schedules, and estimates of market prices have imprecision as well. If the imprecision from each step is ignored in the subsequent step, the final estimate of NPV or profit will appear artificially precise. In reality, there will be far more uncertainty in the final profit estimate – a situation that can result in unpleasant surprises. We’ll explore this further in a future article.

Parting thoughts

It’s easy to feel overwhelmed by the pace of technological change, but as the McKinsey report notes, the potential double-digit gains in improved “precision forestry” management are large and worth pursuing. Hopefully this article and our previous pieces on “Cost + Loss” will help you confidently create a technology strategy for your organization. With this quantitative framework for thinking about the dollars and cents of precision forestry, you can make sure your investments in technology are getting you the maximum bang for your buck.

The NCX Newsletter

Subscribe for the latest thinking and news from NCX.

about the author

Zack Parisa

Zack Parisa

Co-Founder and CEO
Zack Parisa is the co-founder and CEO of NCX. Over the last decade, he has developed and pioneered precision forestry tools that are revolutionizing the way that forests can be measured, valued, and managed. Using satellites, cloud computing, and machine learning, NCX worked with Microsoft to create “Basemap,” the first high-resolution forest inventory of the United States. It is now using this data to build new markets for forest values beyond timber, such as carbon, wildlife habitat, and fire risk. Zack is a forester and biometrician by training. He earned an MFS from Yale University, and a BS in forestry from Mississippi State University.