GPS Accuracy: Part Two

This is a follow up to my previous post on GPS accuracy. Basically, I returned to the same location --the wading pool in the Roosevelt Baths Quad at Saratoga Spa State Park-- and recorded additional sets of GPS fixes using the Garmin GPSMap 64s and the Garmin eTrex. Image One is a screen capture from Google Earth showing the results. The green markers represent the fixes recorded using the 64s and the red markers were recorded using the eTrex. The red dots represent the actual locations.

Image One: Screen capture from Google Earth. The green markers represent locations recorded by the Garmin GPSMap 64s and the red markers are from the Garmin eTrex Venture HC. The red dots show the points where I stood to record the locations.

My intent, as described in detail in the previous post, is to understand the level of accuracy that can be obtained using inexpensive (less than $300) GPS devices and to determine the factors that affect that accuracy. From that perspective, the average accuracy of the fixes recorded on this visit to the Spa State Park site are similar to those obtained on the previous visit. Slightly better overall, but not substantially different.

The majority of the fixes recorded by both units where within ten feet of the actual and both recorded some fixes within five feet of the actual. Using the criteria discussed in the previous post these are excellent results. My methodology is to record a fix with each unit, walk to the next location and repeat the process. Starting at the upper left I proceeded to the lower left, lower right, and upper right corners of the pool. The green marker labeled 001 was the first fix recorded using the 64s and the red marker labeled 001 was obtained a few seconds later using the eTrex. I made three circuits so there are three markers for each unit at each corner.

On the third circuit I used the location averaging feature available on both devices. When using averaging the GPS device grabs a sequence of fixes at the location and averages those fixes to produce a single fix that should be more accurate. I have not seen a description of the algorithm these devices use to do the averaging, but an educated guess is that the units temporarily store fixes obtained using different sub-sets of the full set of satellites from which the unit is receiving signals. Comparing these trial fixes would allow the unit to drop signals that don't fit well with the others. See the previous post for a discussion of conditions that can reduce GPS accuracy.

To make it easier to see how these results compare I grabbed an additional screen capture showing a closer view of the lower left corner of the pool (the second stop on the circuit). And this brings me right back to the title of the first post; GPS accuracy is not a simple matter. As you can see, the distance from the Green marker 010 to the actual is 1.3 feet as measured using the Google Earth measurement tool. That fix, obtained on the 3rd circuit of the pool using the averaging feature of the 64s, was the most accurate fix obtained by either unit. But the 64s also recorded one of the worst fixes of the day at that same location. The green 002 marker, recorded by the 64s on the first circuit of the pool, is 16 feet away from the actual. That's not terrible but it is the 2nd worst result for either device on any of the three circuits. Given that the second time around the 64s recorded a fix about five feet away from the actual (green 006) it was probably the case that the 002 marker incorporated a signal from a satellite with corrections pending, and that that through off the result.

In summary, the accuracy you obtain using inexpensive GPS devices will almost always be better than the 30 foot level that the manufacturers claim for these devices. And it can be much much better with best case results under five feet. But that level of accuracy requires an understanding of how these devices work and what factors reduce accuracy. To obtain these 'best case' results you should turn your device on several minutes before recording the first fix -this allows your GPS device to download the satellite correction data it needs to make proper use of each satellite. Secondly, use the averaging feature of your GPS if it has one. Or, take several readings at the same location a few seconds apart and do the averaging yourself.

Most important, be skeptical. It may be that the one foot accuracy I got on this go round is a fluke. After all, I didn't see that on my first visit to the same site. Leaving me to wonder, what will the next visit show?

Image Two: Screen capture from Google Earth. The 1.3 feet distance from the actual location to the green marker 010 (third circuit) was most accurate fix by either unit. Confouding that, the green 002 marker, also recorded by the 64s, was one of the worst.

Google Earth and KML - Part Two (GETECH)

In the previous post I introduced the topic of working with geographic data in the KML format. As mentioned there, KML is the native data format used by Google Earth (GE) and it is widely supported by both GPS devices and software. I won't attempt to cover all the gory details of working with and editing KML in this series of posts (resource links at the end), but there a few practical things I've learned about working with GPS data in the KML format that I want to share.

As a basis for this discussion I'm looking at two GPS tracks I recorded on a short Kayak trip on the Kayaderosseras Creek near Saratoga Springs. One was recorded using my trusty Garmin GPS and the other was recorded using the GPS receiver in my camera; a Canon Sx260. I'll discuss and compare the accuracy of the two tracks and touch on some other issues you might encounter while using GPS to collect geo-referenced data for use in your projects.

The first issue is evident in the screen shot below (Image 1). The red line represents the track captured by the GPS receiver in the camera. The purple line shows the route recorded by the Garmin. The obvious difference is that the purple line extends beyond the point where the red line stops. That's because the red line stops at the place where I put the kayak into the creek and later took it out (at the Spa State Park Canoe Launch Site on Driscoll Road). I paddled upstream for a mile or so and then floated back down. I had both GPS units turned on and both units recorded the path. The purple line continues because when I got back to the launch point I forgot to turn off the Garmin GPS. So it recorded a track for the short carry from the creek up to the parking area and for my drive home. That data is not useful so the first thing I want to do is remove it.

Image 1: Screen capture from Google Earth. The two lines (red and purple) are GPS tracks recorded simultaneously on Kayak trip on the Kayaderosseras Creek near Saratoga Springs (NY)

I could do that by editing the track directly in Google Earth. That would be easy but it would be extremely tedious. There are several mouse clicks required to delete a point and there are a lot of points to get rid of to clean up this problem. You can see this in image two. For this screen capture I turned off the display of the track recorded by the camera (the red line) and selected the purple track for editing This makes visible the points recorded by the GPS. Each recorded point is represented by a red dot in the image and as you can see, there are a lot of points in just this short section of the route from the creek to my house. To remove this part of the track by editing it in GE you'd click on each points, select Delete, confirm the deletion and repeat. Lots of repeats.

Image 2: Screen capture from Google Earth. With the display of the track recorded by the camera (the red line) off, the track represented by the purple line was selected for editing. This allows us to see the individual points recorded by the GPS receiver. 

A better alternative is to open the KML file containing the track in a text editor and remove the unwanted points from the file. Once you have the hang of it you can edit the data in the file very quickly. There are few steps involved in this process which goes something like this:

1. Export the track to a file. In Google Earth right click on the item in the Places pane, select Save Place As. Be sure to select KML from the format drop down so the file is saved in the KML format.
2. Open the file in a text editor or an XML editor (I use Notepad++; a free and open source editor that is popular with programmers). You may need to configure your editor to recognize that the KML file is an XML file. That should allow the editor display the contents with nice formatting.
3. In this case, removing the points I don't want comes down to removing all the points where the latitude is greater than the latitude of the put-in point. I'm in the northern hemisphere --so the latitude coordinates increase as you go north-- and all the points I want to keep are south of the put-in. Image three shows the map zoomed in on that location with the properties dialog for the point displayed. The properties dialog shows the latitude and longitude of the marker (and the location). This is how I got the latitude I will use as the basis for my edits; any point with a latitude greater than 43.034293 can be removed from the file. A potential complication is that that the coordinates in the KML file are stored using the decimal degree format. That's as you see it in the dialog seen in image three, but that representation is not the default for coordinates in Google Earth. The default is to represent coordinates in the Degree/Minute/Second format; like this: 43°02'03.4548" You can change the representation used by Google Earth using the Options dialog on the Tools menu.

Image 3: Screen capture from Google Earth. Viewing the properties of a marker added at the location of the start/end point of the trip. The properties dialog shows the coordinates of that point.

Once you have a KML file containing your track, and you have a way to identify the point(s) to remove (in this case, latitude > 43.034293) you can remove the unwanted points from the file. To do this you do need to know a little bit about the XML data format and how to edit data stored in XML, but there are many introductions to XML available on-line and I've included references at the end of this post. Also, if you've done any work with HTML this will all look familiar. XML data elements are placed inside of tag pairs and you must delete the entire tag pair or you wil break the formatting of the file (and GE will complain when you try to reload it). You'll also need to recognize how the coordinates are represented. An example from my file looks like this:

<gx:coord>-73.79277999999999 43.034036 102.52</gx:coord>

The coordinate of the location is represented by three values with a space placed between each value (it might be hard to see the space; -73.79277999999999<space>43.034036<space>102.52). The values are:

-73.79277999999999	longitude	a "west" longitude represented as a negative number
43.034336	latitude	a "north" latitude represented as a positive number
102.52	elevation	meters above sea level

As discussed in the previous section, I want to delete any point where the latitude is greater than 43.034293, so this point can be removed from the file. The points are stored in the file in order, so once you find the place break between what you want and what you want to remove you can select all the bad rows and remove them. If editing data in this way is familiar to you then this should get you started. If this is entirely new then you may need more help. Look at the references or ask a friend who knows about this stuff.

With unwanted data removed from your file there is one final step; opening the file in Google Earth so you can see the results of your work. In Google Earth select "Open" from the File menu and find your edited file on disk. If you get warnings about errors check the file over for mis-matched tag pairs. That's it, you can save the file in your Places folder so that Google Earth reopens it automatically or you can reopen the file when you need it in the future.

This post is already rather long so I'm going to cover the other topics I raised at the start in a separate "part two" posting.

Using Google Earth for Geographic Exploration (GETECH)

Readers of this blog know that I use Google Earth (GE) to plan outings and to manage data that I collect using various GPS devices. I use Google Earth as a sort of entry-level Geographic Information System (GIS) and in the following series of posts I plan to share what I've learned about using Google Earth as part of a citizen/community science data management system. I'll probably move these posts out to a separate blog in the future but to get it started I'm putting them here. If you aren't interested in these more technically oriented posts you can skip them. I''ll stick (GETECH) on the title to make them easy to recognize.

On most of my outings I collect geo-referenced data using one or more GPS devices. I have a Garmin hand-held GPS, GPS in my phone and even a camera with GPS built-in. I collect "tracks", the GPS receiver stores a point every few seconds allowing you to trace your route, and I also save WayPoints. These are specific locations of interest and the GPS saves the coordinates along with a name you enter. The Garmin GPS can grab a bunch of coordinate pairs in quick succession and average them to improve the accuracy of the fix. This is useful if you want to get the most accurate location that your device can record, usually around ten feet. I'll do a future post on GPS accuracy to explain this limitation.

After a trip I upload the data from my GPS devices into GE. This allows me to see the tracks (my route) and waypoints in the geographic context provided by GE. If the data was collected for a specific use I move it into whatever storage system I'm using for that project (more on this in a future post). My Garmin GPS is directly supported by GE so I simply plug the GPS unit into a USB port and use the GPS option on the Tools Menu. There are options to allow you select what you want to import. By default GPS data is added into a sub-folder of the "Temporary Places" folder in your GE Places. I typically tweak things a bit and move the imported data into the "MyPlaces" folder. This allows me to save each GPS data set for ready access in GE. You could also right click on your newly uploaded dataset and save it to your computer using the Save Place As option.

It's even easier to grab the GPS data using the phone. I have several GPS related apps on my Android phone (a Google Nexus 5) but I most often use GPS Essentials. As with the dedicated Garmin, GPS Essentials allows me to record tracks and save waypoints (it has other capabilities that I'll discuss in future posts). To get the data into GE I use the Export feature of GPS Essentials and save the data directly to Google Drive (Internet-based storage). It takes a few minutes but once the data shows up in Google Drive I save it to my local computer as a KML file. KML is the native file format of Google Earth and most software that uses geographically referenced data can use data in the KML format.

This has been my basic working process for several years but I've recently become concerned that it was breaking down. The data I collect is best categorized as "ecological inventory". I note where species are found, when I saw them and I record additional information about the circumstances. The "where" and "when" are essential to the data having value and I was not managing this information in a consistent way. It's there --embedded in the KML files-- but I didn't have an easy way to find everything related to a location; or everything for a specific time span. Some of data was stored in raw files and some was in databases that I've created. To maximize the value the data needs to be consolidated and I need a more robust process for recording and managing the metadata. Also, with over 200 top-level sub-folders in my MyPlaces folder, Google Earth was running noticeably slower.

Screen capture of Google Earth after I cleaned up and organized my MyPlaces data. The lines are tracks captured using GPS and the markers are points of interest.

So I looked into how GE actually manages this data and found that all of the data you see in GE in the "MyPlaces" folder is actually stored as a single file on your computer named myplaces.kml. The location of this file varies based on the operating system of your computer and the version of GE you are using. On my computer, running Windows 7 and using Google Earth Pro, the file is saved to:
C:\Users\<your-windows-name>\AppData\LocalLow\Google\GoogleEarth

Seeing the MyPlaces.kml file made it clear why GE was straining a bit. The MyPlaces.kml file on my computer was just under 50 megabytes in size and using a text editor to open it I found that it contained over 1.1 million lines. For context, if we assume that a page of text averages around 60 lines, the MyPlaces file on my computer contained over 18,000 pages. That's a bit much.

I had let the situation get out of hand and now it was going to require a lot of work to straighten it out. One approach would be use the GE user interface. In GE you can save each folder or item in the MyPlaces folder to a separate file on your computer.  You can then delete that item from MyPlaces and open the file when you want to use that data. If the item you save is a folder GE creates a single archive file with the (.KMZ extension) containing the entire contents of the folder. This approach would work but it was going to be a tedious process to say the least.

A second alternative is to edit a copy of the MyPlaces.kml file using a text editor (note that I said "edit a copy" - don't edit the file used by GE). You can use any text editor to open a KML file. Unfortunately, given the size of the file on my computer this also was going to be a slow and error prone process. Using NotePad++ --an editor design for working with lots of text-- opening and navigating the file was extremely sluggish (recall the 1.1 millions lines). I could have copied out sections and worked on pieces of the file but the data is structured (it's an XML file if you are familiar with that sort of thing) and there are references among different sections in the file. It would be very easy to mess this up.

The solution I chose was to write a script to run through the file (a copy of course) and save each folder, document and placemark to a new, separate, KML file. I could then remove (making a copy) the MyPlaces.kml file used by GE (make sure that GE is not running if you do this). The next time GE is run it creates a new and empty MyPlaces.kml file and you are back to a "clean" install of GE. The separate KML files created by the script can then be opened using GE when I need them.

I'm making this script available for anyone who wants to run it and you can access here:
https://gist.github.com/kentstanton/3441cc368d3c52621b19

Please note that you must have PowerShell 5.0 installed on your computer to run the script and you need to know how to run PowerShell scripts. If you are familiar with PowerShell programming you can alter the script as needed. And it would not be too difficult to port the code to a different language if you do not have a Windows computer to run it on. The script requires Windows PowerShell 5.0; the very latest version (as of this writing). I'm working on a larger project that will incorporate this functionality and remove this requirement but I wanted to go ahead and make this available now because it might be useful to some people as is.

Moreau Lake - East of the Hudson -StoneWall Trail

The Hudson River splits Moreau Lake State Park (MLSP) into two sections. The main body of the park --where Moreau Lake is located-- is west of the Hudson River. However, a good sized chunk of park lies east of the river and coming from Saratoga Springs you must cross the Hudson to get there. Confusing the issue is that the Hudson makes a 90 degree turn and flows northeast through the park. So a hiker viewing the Hudson from the Palmertown Ridge is looking west toward the section of the park that is east of the river.

Moreau Lake State Park; The eastern section is on the left in this image of the park trail map.

To reach this eastern section of the park from Saratoga Springs take route 9N to Corinth. Cross the river and drive north for a half mile to the intersection with Call Street.  Turn right onto Call Street. About three miles from that intersection you'll see Hawk Road on the right (one sign says Hawk Street but it is Hawk Road). The first half mile of Hawk Road is paved but you'll quickly reach a steel gate where a sign reads; "Trailhead Parking Straight Ahead". From there it's about a mile on a one lane dirt road to the large parking area. As of October 2014 the road was passable for normal passenger cars with just a couple of rough spots.

The Stonewall trail --with Blue Markers-- leaves the west side of the parking lot (on your right as you drive into the lot). A second blue marked trail leaves the lot on the opposite northeast side. That trail is not seen on park maps that I have but it's a nice addition to the trail system. It makes possible a very pleasant loop hike of about 2.5 miles. The far point of the loop is near the Hudson at a spot where some 19th century building foundations remain. Making the "loop" is a bit confusing because, as mentioned, not all of the trails appear on the current trail maps. Even so, it's easy country to navigate and the continuation of Hawk Road (abandoned) crosses the trail at a couple of points making it possible to return to the parking area by that route.

Through the trees - the Spier Falls reservoir (Hudson River) 

This "Hawk Road Trail" (a name I just made up) also reaches the Hudson; but at a point further downstream just above the Spier Falls dam. From there the trail/road continues to the power line right-of-way leaving the power station at Spier Falls. The trail runs along the right-of-way as part of the canoe portage route around the Spier Falls dam. Canoeists approaching the dam from above leave the water and climb up to the road, follow a short stretch to Hawk Road to the right of way, and return to the water walking downhill under the power line. At that point you are on the upper section of a second reservoir backed up behind the Sherman Island dam about 3 miles further downstream. The hiking trail continues along the east bank of this reservoir for a couple more miles, but the transition between the two sections is heavily overgrown and serious tick precautions will be needed to get through unscathed.

Also worth mentioning is that this entire area is open to hunting and I ran into two hunters on my recent outing there. They politely handled their guns in a way that was safety aware but you should wear bright colors if you hike during hunting season.

As with the better known trails on the west side of the river, these trails offer a wealth of natural and human history to consider as you walk. I got a clear look at a male Red Bellied Woodpecker (Melanerpes carolinus). This species is distinctive and common in many eastern forests, but MLSP is a bit north of its' historic range. Which begs the question of whether this species is expanding its' range northward. And if so, why? Birds are able to respond to changing environmental conditions at a faster rate than many animals and plants (flying comes in real handy). So shifts in bird ranges can serve as an early indicator of changing ecological conditions. If the normal range of the Red Bellied Woodpecker is shifting northward it could be in response to a warming climate. But to make that assertion we need to know the natural history of this species for this area. Has the Red Bellied Woodpecker been present here in the past? If not, when did it arrive and what environmental/ecological changes make the area attractive now?

A less positive indicator of changing environmental conditions is seen in the extensive infestations of Asiatic Bittersweet (AB) along the section of Hawk Road that runs above Spier Falls dam. That AB is present here is hardly surprising. I had previously spotted AB from my canoe; at a place along the river/reservoir bank just below this section of trail. This area has been extensively altered by human activity and once it is present AB thrives where people disturb the forest and create forest edges. AB is common along a half mile stretch of the old road above the dam but I did not spot in the wilder forest areas. It would be interesting to survey the surrounding forest to determine the extent to which it has spread.

Asiatic Bittersweet along the (abandoned) section of Hawk Road near Spier Falls Dam.
Asiatic Bittersweet grows in dense stands that crowd out other vegetation and its' bulk can pull down trees. There is a similar native species; American Bittersweet. The two are easy to tell apart in the fall when the berries are present. If the berries grow all along a stem, it is the Asiatic variety. American Bittersweet has berries only at the ends of stems. Complicating matters is that the two varieties hybridize easily.

To round out my recent day of hiking I took a bush-whacking side excursion from the Hawk Road parking area down to the point where Bennie Brook flows into the Hudson (about 3/4 of a mile). Two things stand out from this little adventure. First, don't try to walk along the edge of the reservoir. It is really tough going and the dreaded deer (black legged) ticks --that carry Lyme disease-- are definitely present. The second point is that I did find something interesting along way; a very large White Oak, diameter 47 inches at chest height. This is probably the largest tree (by this measure) that I've seen anywhere in Moreau Lake State Park. And, because White Oaks are long lived, it could be the oldest as well. White Oaks can live to be 300+ years old and I'd love to know if this tree is really old, having somehow escaped logging, fires and other mishaps. Or is it a younger tree, maybe in the 150 year range, that just happened to grow under particularly favorable conditions. So many questions so little time.

Data: Bittersweet

You can't talk about invasive plant species in this part of New York State and not talk about Asiatic Bittersweet (AB). This vine is one of the most visible and damaging of the invasive species found in the Hemlock Forest. I know a little more about the specifics of this plant because about 10 years ago it appeared in my backyard. I've been battling it there ever since. I cut it off at the ground and I pull it out by the roots. I've fought AB to a draw in my yard but it is everywhere in the surrounding area; especially in a successional field near my house where it has pulled down several trees by the sheer weight of the vines. AB spreads both by the dispersal of seeds by birds (that eat the berries) and via rhizomes that spread out from every established vine. This plant is not going away anytime soon.

Image One: A dense cluster of Asiatic Bittersweet vines in the Hemlock Forest.

Asiatic Bittersweet is not a shy plant and even in a dense forest spotting the vine is easy. It wraps itself tightly around the trunks of trunks of trees and it will bridge from tree to tree in the canopy. But it is in the late fall and winter that AB really stands out. Mature plants produce large quantities of orange-red berries with yellow sheaths that are quite distinctive. The only plant you might confuse it with is its native cousin, American Bittersweet. Asiatic Bittersweet hybridizes with the native variety and by all accounts the hybrid is quickly replacing the native version everywhere that both species are present. Unfortunately, the hybrids appear to take on the negative characteristics of the invader. Namely, it's ability to produce a mass of vines that tend to strangle whatever it gets a hold of. The native variety is certainly not extinct, so before you take actions to control this plant be sure you know which one you are looking at. The plants are very similar in appearance with one easily spotted distinction. When berries are present, if the berries are found all along the vine it is Asiatic Bittersweet. The berries of the native variety appear only at the very ends of each vine.

AB will be one of the plants I focus on in studying the invasive species present in the Hemlock Forest. The general questions I've discussed in other posts certainly apply and, in the case of AB, there is the additional issue of why the non-native variety is so much more destructive than the native species. In the Hemlock Forest AB is found pretty much everywhere there is an opening in the canopy. There are several old fields in various stages of succession in the forest and AB is present in all of them. It is almost certainly present in other places in the forest as well and one of my first goals will be to map the full extent of its' presence. That data can be used in the future to learn more about how AB disperses. AB can also be used to study how community assemblages are changed by a well established and aggressive invasive species.

Map One: The purple polygons show areas where Asiatic Bittersweet is present in significant quantities. This highly aggressive vine is probably found in other parts of the forest but I have not yet specifically mapped its' presence.


References and Resources:
Asiatic Bittersweet (Wikipedia)
Mistaken Identity: Invasive Plants and their Native Look-alikes

Base Maps

In earlier posts I introduced the Hemlock Forest area and I talked about why I want to understand how ecological communities are changed when non-native species become fully established (look back at those previous posts for the details). I've also said that anyone with motivation can acquire and contribute valuable scientific data. What I have not talked about is what data you might get and how you might get it. The what and how will vary based on your interests but the thing I can do is to describe my project in detail. If you extrapolate from there.

As for the "what" you have to start with a question; something that interests you. I want to understand how non-native species alter existing ecological communities and species assemblages. My thinking is that in many cases these aggressive non-native species are not going away. They are here to stay. So, at what point do the invaders alter previously existing communities to an extent that a new community type is present. Most importantly, are those new assemblages of species more or less fit in terms of their capacity to provide ecological services and to resist future disruptions. It's commonly held that the general reason invasive species are so successful is that predators and pathogens that co-evolved with the invader are not present in a new locale. That is certainly part of the story but I wonder if some species are in effect "generalists" able to dominate where ever they are found. Take Japanese Knotweed for example. Once established, you can repeatedly mow a stand of Japanese Knotweed off at the ground for several years running and it will grow back with vigor as soon as you stop. And, if you are not paying attention, it will spread to surrounding un-mowed areas via rhizomes. Eventually you can kill it by depleting the energy reserves present in the roots, but Japanese Knotweed is one tough plant. Human activities have increased the speed at which plants like Japanese Knotweed are dispersed but it seems possible that this plant has evolved strategies that would allow it to spread far and wide on its own. We need to better understand how these aggressive generalist species operate because it seems likely that they will eventually be present globally where ever there is suitable habitat.

Back to the project, the first thing I want to know is which communities are present and where they are located. This is the easy part for me because my interest in this has emerged from working with geographic information systems (GIS). A GIS allows you to place your data into a geographic context. Using a GIS you can visually look for patterns in the distribution of the phenomena of interest. Going further, there are a wide variety of analysis techniques you can use to learn more about the phenomena as represented by your particular set of data. In short, I am starting the project with the creation of a GIS.

The first thing you need are base maps. As a term, "base-maps" doesn't have a distinctly defined meaning. It refers to whatever map layers are needed to put your data into a geographic context. These days we can download a wide variety of base-map layers from the Internet. Believe me, it didn't used to be that easy. And, if you are working outside of the United States, it might still be a major undertaking to get the base-map layers that you need. For my project I'm starting with a raster base-map layer created from a USGS topographic map. The USGS refers to these as digital raster graphics or DRGs. I'll be writing about where you can get map layers of various types in future posts but, for now I'll let the images tell the story.

A digital raster graphic (DRG) base layer showing the hemlock forest. The red line delineates the boundary of the land that is included in the Saratoga Spa State Park. The orange line shows the marked trails coming from the parking area on Crescent street (upper right in the image). Notice that in the lower left portion of the image the red boundary line does not follow the boundary visible in the underlying DRG. The boundary line was digitized from a map of the State Park and there is a slight discrepancy between the two sources. This could be the result of the two sources using different surveys or it could be a cartographic discrepancy coming from differences in the scale and accuracy represented by the two different sources.

I have also acquired aerial imagery of the Hemlock Forest Area; orthoimagery made available for public use by the New York State GIS Clearinghouse (see the Mystery Lines post for more information on this imagery source). The images are geo-registered and line up pretty well with the underlying topographic map layer. My data will be layered on top of the background provided by the DRG and the orthoimagery and I've started by creating a layer containing the park boundary and the state maintained trails present in the forest.

Image Two: The Hemlock Forest aerial image base layer with park boundary (red) and trails (orange) superimposed. Many details emerge from this view of the forest. For one, the old growth hemlocks that provide the area's namesake are easily seen clustered in the area circled by the upper trail loop.   

My interest is in ecological communities so I need to know what communities are present and where they are located. That information is available in a map published along with the Spa State Park Master Plan released in 2009. The ecological communities map is provided as an image in a PDF file. You can't add this PDF to the GIS directly but I was able to take the image and register it with the base maps in my GIS. I'll document the process used to do that in a future post but images three and four show the results A key feature of the GIS software is that it allows you to control the level of transparency used to display each layer.

Image Three: The ecological communities map registered in the GIS. I grabbed the image from the PDF using standard screen capture software. The screen capture was saved to a JPEG format image file. That file was added to the GIS and registered with the base-map layer. As seen here, the ecological communities layer is displayed as a non-transparent layer; you can't see through it but you can see the base layers around the edges. The boundary and trail outlines are visible because that layer is displayed on top of the eco-communities layer. Each colored area represents a different community type and the zones can be interpreted using the legend found in the original document. 

Image Four: The same map with the eco-communities layer set be partially transparency. Now we can see through the eco-communities layer to the underlying imagery. Using the GIS in this way each additional layer provides more information and seen in combination patterns that were not previously visible become apparent.

There's one more layer that I need for may base layer set. That is a layer of just the outlines of the community zones created in a way that will allow a database table row to be linked with each zone. This linkage between map objects and data is what GIS is all about and this capability will provide the basis for future analysis. I created the layer by "hand digitizing" the ecological community image; essentially I traced the outlines of the zones using tools provided in the GIS. The window seen in Image Five shows data associated with the object selected in the map (the blue area in the center). I'm just getting started, so I have only a couple of columns in the database table. But now that the linkage is in place I can add more columns as I need them.

Image Five: The complete base-map set from the GIS. The black outlines are the zonal boundaries for the various ecological communities identified in the Spa State Park master plan. Each outlined object is linked to a row in a database table. This allows the map representation of each ecological community to be associated with whatever data we want to capture.

Reference and Resources:
The Spa State Park Master Plan Eco-communities map

Where is this place? The Hemlock Forest

Where is this place I've been calling the Hemlock Forest?. The map below should help. Open up Google Maps and view the Saratoga Springs (NY) area (terrain view is best). The park is a prominent feature south of the city.

Map from Google Maps. The Hemlock Forest area is the park area (darker green) on the right. The orange lines show the trails and the marker is at the location of the parking area.

Of particular interest to me are the ecological communities present within the park and surrounding areas and the 2009 Park Master Plan includes a map of Park's ecological communities (You can view it here). Overall, Spa State Park is more developed than many others, but it also incorporates forested areas and a number of significant wetlands including an unusual Perched, Swamp White Oak, swamp.

The designation of ecological communities is based on both the landscape and the plant and animal communities that are present. The New York State Natural Heritage Program (run by the Department of Environmental Conservation) maintains a list of the community types found in New York State and where examples of those communities can be found. It's worth noting, however, that the classification of ecological communities is far from cut-and-dried. That's not a criticism, both the program and the website are well done and, for me at least, fascinating. But these designations are open to some interpretation and the communities change over time. It may be that new community types will have to be defined as natural systems adapt to changes resulting from human activity.

The potential for undesirable change is raised as a concern throughout the Park master plan which calls out the presence of invasive species -primarily plants and insects- as a significant threat to the natural heritage of the park. Several aggressive invasive wetland plants are present in the Hemlock Forest area (as they are throughout the park) and these plants pose a threat to native ecological communities. Two highly aggressive non-native plants -Oriental Bittersweet and Japanese Barberry- are now the dominant plant species in several areas of the Hemlock Forest. I plan to map these locations as part of surveying the state of the forest. Japanese Knotweed, another highly aggressive invasive species, is present near the parking area and that's worth watching. Once Knotweed gets going it is incredibly difficult to control.

Additional resources and information that you might find to be helpful.
Saratoga Spa State Park Master Plan
Saratoga Spa State Park Master Plan Map
Saratoga Spa State Park Master Plan Ecological Communities Map

Mystery Lines: What are those lines seen in the satellite images?

I love to visually explore an area using Google Earth (or similar software) and then go off and match up what is actually present on the ground with what I could see on the computer. Google Earth provides easy access to high resolution images but, if you are willing to dig a little deeper, there are many other sources of spatially referenced data that can be analyzed to learn about areas we want to study.

But let's talk specifics. Looking at satellite images of the Hemlock Forest area I noticed that there are some odd looking parallel formations present in one area of the forest. You can just make them out in the image from Google Earth (Image One). What are they? What process created them?

Image One: The lines are seen in the center of the image running at a 45 degree angle from upper right to lower left. The straight line that runs from top to bottom is a stone wall. Source: Google Earth, Historic imagery from April 1997.

Viewed in Google Earth I couldn't see enough detail to come up with a plausible explanation for what the lines represent or how they were created. But, for locations in New York State, other options exist including the very high resolution orthoimagery available from the New York State GIS Clearinghouse (NYSCG). Orthoimagery is a fancy name for aerial photographs; pictures taken from airplanes using high resolution cameras and techniques that minimize distortion. When you take pictures of the earth from a satellite the distance between the satellite and the ground pretty much ensures that the camera is at a right angle to the ground. So there is little or no distortion of the image. When you take those same pictures from an airplane, if the camera is not perfectly parallel with the ground, the images are distorted.

The imagery available from the New York State GIS Clearinghouse (NYSCG) covers (nearly) all of New York State and is available to the public via the NYSGC website. The catch is that there are thousands of files and they are big files. And, once you find the right images, you need a way to view them. These images are geo-referenced meaning that using the right software you can view the images and they will line up correctly with other map-based data layers. The software you use is commonly referred to as GIS; an acronym for Geographic Information System. Loaded into a GIS these geo-referenced images can be analyzed using a variety of methods. They can also be used to provide a base layer on which you overlay other data. This is much like what you do when you view satellite imagery in Google Earth and then add points or lines or outlines (polygons) on top.

The Hemlock Forest orthoimagery (Images Two and Three) (viewed using Quantum GIS) reveals a lot more detail and those curious lines now take on a distinctly man-made look. Particularly telling is the track that comes in from the left and crosses the formation (Image Three). They look like tire tracks but when taking measurements in the GIS I found that the lines in main set are about 50 yards apart. That made me think that they might be ditches created using a tractor or bulldozer. It is also more apparent from the orthoimages that the formations are filled with water. The dark color seen in the formations is the same as the color of small ponds and streams present in nearby areas.

Image Two: The lines are in the center of the image running at roughly a 45 degree angle to the top of the image. The red lines represent the boundary of the State Park. The park boundary is a separate layer viewed on top of the images in the GIS. Source: New York State GIS Clearinghouse.

Image Three: Zooming in for a closer view the structures and the "track" coming in from the left have a distinctly man-made look. Source: New York State GIS Clearinghouse

Still, I had to wonder why someone would create ditches -about 50 yards apart- through the forest at this place. That left only one solution: field trip. From the Hemlock Trail parking area twenty minutes by trail and another half hour winding my way through the swampy forest brought me to the place pictured in Image Four. This, clearly, was one of the ditches seen in the orthoimagery and it was also immediately apparent that these are ditches that someone dug.

Image Four: The ground view of one of the water courses seen in the aerial photos. Filled with debris and leaf litter these were probably a few feet deep when created. The surrounding area is a swampy, nearly level. The entire area has very poor drainage. So it seems likely that someone dug these ditches to provide drainage for some type of agriculture.

The ditches are roughly 50 yards apart and run in straight lines through the forest for two to three hundred yards. What was not apparent from the aerial imagery is that the area where these ditches are found is very flat and very swampy. A few larger and older trees are present but most of the trees are roughly the same size (and age). Based on the size of the trees I estimated that this area was probably in use for some type of farming 50 years ago.

But something was still missing. The surrounding areas were all clearly used for agriculture in the past but this area is very wet and it is surrounded by swamps. My visit took place in late November and it had been generally dry, and even so, every depression in this area is filled with water. During wetter parts of the year this area would likely be flooded. The trees that are present are types that you find in a hardwood swamp or floodplain forest.

However, after poking around a bit more, a possible explanation appeared. One of my interests is in how invasive species compete against native species so I'm always on the lookout for invasive plants and trees. And I noticed the presence of a number of small, stunted, trees interspersed among the common species. Being late November, with the leaves down, I couldn't tell right away what those trees were. Then, one of them gave up its' secret, on a branch I spotted a small apple. The stunted trees are apples trees and now I could see that someone had tried to drain the area to create an orchard.

It appears that that plan did not work out very well. Apple trees prefer well drained soil and perhaps the farmer was unaware that this area is notable for lying over a layer of impermeable clay. Not too far away that clay, combined with the nearly flat topography, has produced a type of wetland known as a "perched" swamp. A perched swamp is a wetland where the water on the surface is not connected to the local water table but where the drainage is so poor that you get a swamp anyway. In most swamps the water level rises and falls along with the local water table but this is a place where the water doesn't drain through the soil and runoff is slow or non-existent. Creating an orchard here was never going to be easy. Which may be why we now have this wonderful little wild area to explore and enjoy.

Resources:
New York State GIS Clearinghouse Orthoimagery

A little Wild: The Hemlock Forest

Things are changing. The way people think is changing. The natural world around us is changing. Everything is changing. This has always been the case; change happens. But one thing is different now. We humans are altering the global environment at a pace and scale that from the perspective of geologic time looks a lot like the great natural cataclysms of the past. Think in terms of huge super volcanoes that shook entire continents or the asteroid that killed of the dinosaurs. We are the human asteroid. We are in effect running a giant experiment on how the earth reacts to change at scale and lots of data is coming in. Our instruments and satellites capture immense amounts of data every day. Leaving us with the questions: "what can we learn?" and "what should we do?". That's the context for this blog.

OK, setting that cheery opening aside, I'm planning to write about very specific places that I think are interesting. One of those places is a small (300 acres) forest that is part of Saratoga Spa State Park. I call it the Hemlock Forest though that name is completely made up. Before going any further, let's look at a map (Figures One and Two).

Figure One: The City of Saratoga Springs, New York and surrounding areas. The green outline represents the boundary of Saratoga Spa State Park. The Hemlock Forest is on the right wedged in between Rte. 9 and the I87 (the Northway). The park covers approximately 2800 acres and that includes a tree nursery two golf courses, a Hotel and an outdoor performing arts center that seats 20,000 people (SPAC).

Figure Two: The Hemlock Forest is outlined in red. This section of Saratoga Spa State Park covers about 300 acres. The forest is separated from the main body of the Park by Rte. 9 (on the left). The northern edge of the forest runs into the rapidly developing urban/'suburban fringe of the City of Saratoga Springs. The southern margin is mostly undeveloped with a connection to the Kaydeross Creek other nearby natural areas.  

There are several interesting things going on in this little bit of forest. One is the presence of wetlands of several different types including an uncommon "perched" swamp (perched referring to how the swamp interacts with the local water table). This particular wet area, though small, hosts an uncommon ecological community anchored by the presence of old growth Swamp White Oak trees. The forest also includes an old growth hemlock forest and good stands of American Beech trees that so far have held up against the Beech Bark Disease that has devastated Beech trees elsewhere. The State Park maintains a two mile long trail that loops through a 150 year old forest that includes hemlocks, White and Red Pines, and Oaks. The existing trails provides direct access to less than half of the total area though the un-trailed southern portion of the forest is hardly an untrammeled wilderness. The entire area is crisscrossed by stone walls, old farm and field tracks, and unofficial trails. Still, parts of the forest have a distinctly wild feel and the range of habitats present make this an important natural preserve.

I enjoy walking the area and capturing data that represents the "current state" of the ecological communities that are present there and that will be one of things I'll write about in this blog. This "Field guide to the Hemlock Forest" will also include information on technologies and methods that I use, and that anyone can use, to gather real data about places like this. I'm sure that some posts will start with something like "saw a  Barred Owl today", but my emphasis will be on acquiring and managing data that represents the ecological and environmental state for this area. My hope is that this data will be useful in the future. In particular, I'm interested in how invasive species interact with native communities and I'm hoping to capture a data set that can be used to learn more about how those interactions change over time.

When I was a kid I loved to read about the explorers and naturalists who traveled the world finding species not known to western science. Unless you are willing to spend a lot of time looking through a microscope it's unlikely that any new species will turn up in northeastern New York, but spotting the arrival of a species that has not been present in the past is a very real possibility. We have tools at our disposal that were unimaginable even 20 years ago. A smartphone with GPS, environmental sensors, and on-line identification guides doesn't match a tricorder quite yet, but that's where we are headed.

If you are reading this it might be because you live nearby and can visit these places. Or maybe you are interested in nature generally and my writing is not so bad as to ruin it for you. Or, maybe, and this is my hope, you want to do something like this where you live. I'll be writing about citizen science and how we can contribute data towards a better understanding of our world.

That's all well and good but you may get an inkling that the for me this is fun. The Forest is less than a mile from my house and I like to walk in the woods. I used to play golf but, to paraphrase Mark Twiain's famous quip, "that's a sure way to ruin a nice walk." There is a parking area with trail access on the northern edge of the forest (access is from Crescent Avenue). The trails are lightly used and I'm yet to see another person while wandering off trail. Much of the area is swampy so if you do go off trail you can expect wet feet. And I'm not worried that publicizing the area will bring in swarms of hikers. For much of the year the hemlock forest is well defended by armies of mosquitoes and ticks (beware: Lyme Disease, seriously I have captured many many Deer Ticks while walking in this area). For now the Hemlock forest is well protected.

Resources:
Saratoga Spa State Park Master Plan (pdf)
Saratoga Spa State Park Master Plan Map