I can’t really hide it: I’m pretty excited to be writing about this program. I understand that that may sound a little odd. I mean, who would be excited about a program named ForWarn II, especially given the slightly ominous sounding name?
Here’s the thing: I’ve been looking for a program like this for a long time. I’ve long loved spending time in nature, but in the last few years I’ve finally made a serious effort to make sense out of it. As I live several hours from the nearest true wilderness, I’ve had to resort to a lot of “remote learning” when trying to make sense of forests and everything in between.
This has been a bit of a struggle, as each tool or piece of software always left a few things lacking. ForWarn II (I’ll just refer to it as ForWarn in the rest of the post) has all of the capabilities I’ve been looking for and much, much more.
What is ForWarn II and Who Uses It?
This section will be brief, but it’s important to give a bit of a background before we dive in and start playing with the bells and whistles.
Publicly Available Software Designed to Detect Changes in the Natural Environment
This software was put together by the U.S. Forest Service (a part of the United States Department of Agriculture), and it is designed to aid in the discovery and tracking of all sorts of things that impact nature: natural or not.
From tornado damage to wildfires to invasive insects, ForWarn II uses satellite imagery to detect changes the vegetative characteristics of our forests in the United States. In addition to monitoring our forests for changes, this software pulls in a bunch of other sources of complementary data.
Appears Mostly Used by Career Scientists
I can’t be totally sure, but I’m pretty convinced that the vast majority of ForWarn users are career scientists. There doesn’t seem to currently be much of an appetite for it in the mainstream, which is a bit of a shame but totally understandable. There’s no way around it: there’s a lot of functionality built into this software, as it appears to be a “jack of all trades” tool designed for a wide variety of scientific purposes.
Why Should I, a Normal Human Being, Use This Program?
I get it. Maybe you’re like me and you just happen to like anything and everything to do with nature. That’s great! So why should you care about using some tool that was designed by and for scientists with a bunch of letters behind their names?
Simply put, because with a little knowledge and practice, ForWarn is a tremendously capable tool, especially for those of us who love nature and want to spend more time outdoors.
Awesome Things are Possible With This Tool
Here are some of the things that are easily possible with this tool:
- Finding massive colonies of wild ramps in the spring (seen from space, no kidding)
- Locating groves of northern red oak trees, as seen in the fall when their leaves change to a bright red color
- Quickly finding National Forest land clear-cut in the last few years, in order to find a potential new raspberry and blackberry spot
- Locating stands of a specific kind of tree I’m learning to identify as I’m seeking to understand the forest around me
- Assessing the damage from a recent tornado or windstorm, in order to stay out of the way of the clean-up effort
And so on. If you’re a forager, I’m hoping I now have your attention. Even if you’re not into the whole wild foods scene, the ForWarn tool is an incredible opportunity to start to make sense of the nature that surrounds you. For me, my efforts to make sense of nature has been an amazing journey and I’m extremely grateful for all the ways that the ForWarn tool has aided.
So while the truth is that there’s a bit of a learning curve here, I think there’s a lot of value that the average person can get out of it.
Quick Note: What is NDVI and Why it Matters Here
Before we get into the fine details of how to effectively use ForWarn, there’s one matter we should briefly cover: what on earth is NDVI?
In layman’s terms, the Normalized Difference Vegetation Index (NDVI) is a method of determining how healthy and productive the plants in a location are at a given time. This is done by analyzing satellite photos for certain parts of the light spectrum. Basically, healthier plants more effectively use photosynthesis to convert certain lights into chemical and physical growth.
Healthier plants have a higher NDVI score, and a low score indicates either a dead or dormant plant. For example, a deciduous tree will have the highest NDVI in summer and the lowest NDVI in winter. If you’d like to learn more about the mechanics on how NDVI works, check out our post that goes into greater detail.
How ForWarn II is Setup for Use
There’s no way around it: ForWarn can be a bit overwhelming when you first open it up. This is about what you should expect when you open up ForWarn for the first time:
You can see that there are really four unique parts to this interface:
- There’s a navigational bar running across the top of the screen featuring a variety of buttons and functions
- On the left there is a ‘Map Layers’ menu; layers are superimposed onto the base map when their box is checked
- On the right you have a ‘Map Tools’ menu
- The main frame has the base map specified in the navigational bar and a layer superimposed over the top of the base map
So, let’s dive in: not all of the options present will be useful for our purposes, so don’t worry if we appear to skip over some features.
The Top Toolbar Has Some Important Functionality
Here are the main things you need to know about the top toolbar.
The Info Button Gives You Data From Map Layers
The info button is the text bubble with the letter ‘i’ that looks like this:
This will likely be the toolbar button you use the most, as this is what allows you to get the exact data from a square you click.
Here’s an example of what I’m talking about: there’s a map layer that I use very frequently that shows the forest type of your region. Here’s an example of what that layer looks like when loaded over a forest:
Each of the colored squares represents a certain type of forest. If you head over to the ‘Map Tools’ menu you would find a key in the ‘Legend’ drop down that specifies the type of forest associated with each color:
As this layer depicts 132 different forest types, it’s extremely difficult if not impossible to understand which shade of red on the main screen represents what forest type.
So in order to tell what each square represents, you’ll need to toggle on the info button. Once the info button has been clicked, clicking on the map will bring up results like the following:
So now we know that the square clicked represents black spruce trees.
Not every layer will have valid data, as sometimes you’ll get null values returned.
Theme Drop-down Menu Controls Your the ‘Map Layers’ Menu Options
The ‘Theme’ drop down menu controls the options that are present in the ‘Map Layers’ menu, as no theme shows all possible layers at one time.
I prefer to leave it on the default theme, which is the following: N. American Vegetation Monitoring Tools. This has all of the layers that I use most frequently, so I just leave it be.
The NDVI Graphing Button: Awesome but Best for Experienced Users
You can find the ‘NDVI Graphing’ button by finding the following icon on the toolbar:
Technically, the purpose of this button is simple: it brings up the NDVI graph for the place on the map you clicked. Here’s an example of one such graph for a forest:
What does this graph mean? In so many words, this represents the previously discussed NDVI values for this piece of land since 2003 (in 8-day increments). If you remember correctly, the NDVI basically indicated how densely vegetated with green plant matter this land is at any time. As you might be able to guess, the great swings in NDVI values each year indicate that this is a deciduous forest.
On the other hand, a dense forest of evergreen trees is still variable in NDVI, but the difference is less stark:
Now here is the bonus part on why this chart is incredibly cool in my book. Notice how there’s a noticeable drop-off in the NDVI peaks between the years 2019 and 2020? A severe storm came through this area in July 2019 and knocked down a bunch of trees in this area. With a certain percentage of trees lost in this area, it produced a temporary drop in the vegetative performance that is clearly measurable.
Here’s an even clearer example: this is a deciduous part of the forest that was utterly devastated in that same storm, and therefore that land was clear-cut in an attempt to salvage the downed timber:
Notice how severe the drop-off is in-between the 2019 summer and the 2020 summer.
This makes sense if you think about it: a mature forest with a dense canopy has a much greater ability to produce green vegetation than newly clear cut land. Yes, the clear-cut land will rebound in no time, but when you compare the productivity to the former forest it has a ways to go.
Everything Else on the Toolbar is Self Explanatory
The rest of the menu is either self-explanatory (zoom in, zoom out, etc.) or mostly unnecessary for our purposes.
The zoom-in, zoom-out, and pan buttons all work as to be expected, but most people will likely just their mouse buttons and scroll wheel. The map behaves just the same as Google Maps, so feel free to scroll and drag to your heart’s content.
The ‘Base Map’ drop down menu is also simple to use. The menu features eight different options to use as a base map, so feel free to use whatever type of base layer you like. I believe the the ‘Imagery’ layer is just the Google Maps satellite image layer, so I typically use that.
The ‘Map Layers’ Menu Shows the Data Sets Available
I’ll just come out and say it straight away: the ‘Map Layers’ menu has an intimidating amount of layers available. There’s just so much here, that it would be easy to get overwhelmed quickly.
In other words, I’m only going to discuss the layers that might be of use to the average person, and we’ll leave the rest behind. Like I mentioned above, the ‘Theme’ drop down menu in the toolbar changes what data layers are available. I’m going to stick to the first theme mentioned in the drop down, as it contains all the information I’m interested in.
So, what exactly am I interested in with this menu? I break it down into the following four ideas:
- Accessing near real-time satellite imagery
- Looking for severe changes in NDVI values that might indicate storm damage or logging
- Historical accounts of U.S. Forest Service logging activity
- Plots that indicate which types of habitats or trees are where
I’m also interested in the Phenological Regions group of data layers, but I don’t necessarily know how to use the information yet.
Before we get into each of the ideas I mentioned above, here’s how you can adjust the transparency on any layer. For the layer you’re adjusting, click the wheel to the right of the layer name:
Then, adjust the transparency by moving the slider highlighted below, and the main screen will updated behind you:
Nearly Real-Time Satellite Imagery: a Major Win
This was one of the things that I was most excited about when I found out about ForWarn. I’ve been looking for a decent set of near real-time satellite images for awhile now, and here we have a few options.
First things first, you’ll want to use the ‘Planet.com Imagery’ section if you’re in one of the following states: IL, IA, IN, KS, NE, OH, WI, ND, and SD. I’m not sure why only parts of the Midwest have access to this layer, but I’m not complaining. Don’t get me wrong, there are other valid options for everyone else, but my impression is that this layer is of the best quality.
Everyone else should check out the ‘High-Resolution Sentinel Imagery’ section, as that provides a decent satellite image for the Continental U.S. You’ll want to select the ‘True Color’ option, as that depicts the colors as seen through human eyes.
Finding Extreme Swings in NDVI Values
This is probably the coolest functionality that ForWarn is capable of, but it’s also rather confusing when you’re first starting out.
Here’s what you need to know: ForWarn is ultimately best known for its ability to detect departures in vegetative performance when compared to past records. What does this mean for us regular people in real life? There are two scenarios that I think are most useful for us:
- Graphically depicting the impact of real-life events (tornadoes, storms, etc.) on our natural surroundings
- Allowing us to identify locations undergoing an event like clear-cut logging
Why Be Interested In These Graphical Depictions of Natural Events?
I have two main reasons why I find this feature useful. First, I think it’s just plain cool to be able to graphically depict the impact of a major event like a windstorm. Second, navigating the woods after a major storm is a complete nightmare. Sure, you can get an idea of the level of damage by driving around, but it’s very difficult to understand which parts of the forest were most severely impacted.
With this tool, I’m able to quickly pull up a map that shows me where the storm had the greatest impact. As I’m looking to avoid the areas severely impacted for safety reasons, this allows me to keep enjoying our forests without unnecessary risks.
So how do we make this happen? What we’re looking for is the ‘ForWarn II Near-Real-Time Change Maps’ part of the menu. Expand that you’ll see a wide variety of options.
Here’s more or less what you need to know about these layers: the goal of these layers is to depict the changes in NDVI numbers when compared to previous years. In other words, we’re looking at the relative increase or decrease in vegetative production when compared to the same time of year in previous years.
The different options present here in the layers menu represent the different time frames over which you can compare numbers. For example, the first group is the ‘From Prior Year’ group, and this just compares the current NDVI values to this time last year. Scroll down through the layers and you’ll notice that there are many different options for controlling both the time frame and the type of data to compare (median values, 90th percentile, etc.).
As we’re looking to measure the impact on mature forest from a major event like a tornado, it likely makes sense to use something like the ‘From Prior 10-Year 90th Percentile’ layer. Using this layer allows you to measure the current vegetative performance against the best years in the last ten years.
Doing just that, here’s what a forest can look like one year after a major windstorm that levels hundreds of thousands of acres of mature forest:
As you can probably guess, red means a large drop in vegetative productivity, while blue means an increase.
Therefore, I would be able to use this layer to find the degree in which the different forests around me were impacted by this storm. This saves me loads of time and allows me to focus on the parts of the forest minimally impacted.
Why Would I Want to Find Areas Recently Clear-Cut?
While I understand when people are upset about the clear-cutting of a forest, it’s undeniable that a clear-cut forest undergoes a very interesting transition in the years to follow.
As someone with a passion for foraging, I’ve come to really appreciate forests at all stages of growth. While a mature grove of oak trees can produce a remarkable acorn crop, recently clear-cut land is a gold mine for all sorts of interesting foods.
Therefore, I’m interested in locating parts of the forest that have been recently clear-cut, as I’m interested in how they look in a few years. Here are the steps I use to find these areas:
- Look for areas with sharp drops in NDVI levels (we’re looking for dark red on the map)
- Toggle off the NDVI layer and ensure that the Base Map is set to Imagery
- Pull up the near real-time satellite imagery to see how it compares to the base map
As you can probably guess, there’s a lot that could be said about this, so if you’re interested in learning more, you can head over to this post which covers this topic in detail.
Detailed Logging Activity from the U.S. Forest Service
Unfortunately, this will only be useful for those that are researching National Forests, but it is still well worth mentioning.
The U.S. Forest Service has more than 20 years of historical logging records available in ForWarn. When the data displays correctly, you’ll have access to the following kinds of information about the logging activity in a particular spot:
- What kind of logging was performed (commercial thinning, clear-cutting, etc.)
- When the logging was complete
- Which forest the logging was done on
- The acreage of the parcel logged
In my experience, not every logging data set from the Forest Service correctly displays this information, but many of them do.
As you may have guessed, if you want to find the details associated with a logging act, you’ll have to turn on the ‘Information’ button on the toolbar and then click on your parcel. That will bring up a screen similar to the following:
For my purposes, I’m mostly looking for land that has been clear-cut. This is due to the fact that clear-cut land allows for the most light access for new plants. As such, I’ve had good luck foraging on land that has been clear-cut in the last 10-15 years.
So where can you find these data layers in the menu? Find the ‘Additional Assessment Maps’ menu item and expand it. Scroll past the first group of layers, where you’ll find the ‘USFS Loggin Activity’ group of layers. This is where you can toggle on individual logging years, or to use the combined option that is listed first.
I prefer to use the individual years, as they show the exact outlines of the locations logged, as opposed to a pixelated map.
Information on Habitat Types and Trees Present
I think one of the most valuable aspects of ForWarn is the ability to pull in information about the expected type of habitats and plant species present in an area. Like the example in the part of this post that explained the ‘Information’ button: there we were able to determine that the land had a value of ‘Black Spruce’ assigned to it on the Forest Type layer.
And here’s the tough part to swallow: it’s likely obvious to you that this type of information is a great over-simplification of what’s actually going on with that piece of land. That’s totally true, as I believe that each pixel of a layer covers about 13 acres of land. But even with this information being an over-simplification of the actual reality, there is still a ton of value we can derive from it. After all, remember the old adage: perfect is the enemy of good enough.
What do I mean by that? Even though I know that that entire parcel won’t be black spruce, I can now reasonably ascertain that I would be able to find black spruce trees if I walked to those coordinates and started searching the land around me. Considering that there are hundreds of millions of acres of forest in the United States, to be able to quickly get in the ballpark of a certain tree without wasting significant time is a very valuable place to be.
Where can you find these layers that show habitat information? They are also in the ‘Additional Assessment Maps’ part of the menu, and the most valuable layers are near the bottom in the ‘Landcover’ group of layers.
In these layers, you’ll most likely be interested in the following options:
- Forest Type
- Major Forest Group
- NLCD 2006
- 2001 GAP Landfire
- Land Cover 2010
- LANDFIRE Vegetation
Feel free to turn the different layers on and off, while checking out what information they contain by using the information tool. Basically, these different layers offer slightly different ways to make sense of the land via classification.
Some layers break down the land into individual tree species (Forest Type, for the most part), while other layers assign predefined habitat types (LANDFIRE Vegetation). I think the point here is to play around with it a little and see what you end up liking.
The ‘Map Tools’ Menu Offers Additional Functionality
Much simpler than the ‘Map Layers’ menu, the ‘Map Tools’ menu has a few important functions.
Here’s the un-expanded view of what this menu looks like:
There are only three things to use here, so this section will be pretty straightforward.
Filtering Out the Noise With the Masks Section
First, we’ll look at the Masks section, which is designed to be used in conjunction with Map Layers from the ‘ForWarn II Near-Real-Time Change Maps’ section. Here are the options in the expanded Masks menu:
This does pretty much what you’d expect it to do. When you turn on a Map Layer that shows the change in NDVI values over a certain time period, you’re allowed to filter to show only certain land types. For example, here’s a map showing the change values for only the grasslands in a heavily wooded regions:
As most of that screenshot consists of woodlands or croplands, there’s very little data to show. However, if you switch it over to deciduous forest then your screen show’s many more values:
Sharing a URL for Your Exact Map, Layers and All
Next up, the ‘Share this Map’ allows you to do just that. There’s a text box where you’ll be able to copy a URL that allows you to share an exact copy of the map you’re looking at with anyone.
You can notice the URL change anytime you make a change to the map.
Legend Conveys Information and Allows for Layer Maintenance
Last but not least, the legend section has two main functions.
First, the main function is to provide a key for the data present on the main screen. Here’s an example of the information available:
There’s one more thing to be aware of regarding the legend section: clicking on the legend for a layer will actually toggle the layer off on your main screen.
I’ll be honest, this took some getting used to on my part, but it is quite handy. Think of it like this: if you didn’t have this then you would have to dig through the massive ‘Map Layers’ menu to find the few layers you have toggled on.
Conclusions and Next Steps
If you’ve read through the entire post it’s more than possible that you’re a little overwhelmed. That’s more than OK, as this is a lot to take in. With that being said, you don’t have to understand everything right away, and this post was designed to merely give you a taste of what is possible.
Don’t worry, the more you use this the more it will make sense. I’ll be incorporating this program on many posts on this site, the difference being that their I’ll be much more practical.
If you’re interested in seeing a more practical usage of this technology, check out this post: How to Find Forests That Were Recently Logged. In this post I go into great depth on how we can use this application to not only find past logging activity reported by the U.S. Forest Service, but also how to identify logging occurring in real-time.