Brandon Smith (Redwoodtwig)

• Draw a map of the place and divide it into areas.

        For each area, identify 1 or more features, natural or human built

• Map in hand, walk property and locate enough observation points that panoramic photographs will cover the entire property, after the leaves have fallen.

        Set up permanent markers at each observation point

        Capture 360 degree panorama photographs from each observation point

        On a copy of each panorama annotate features, guilds and plants that are to be itemized in the inventory.

• Determine the information to be collected at feature, guild and plant levels.

        Create a database in which to store itemized details of the features, guilds and plants that the inventory will track in detail.

• On a clear day in winter, take camera and tripod to each observation point and take the photographs for the panoramas.

       Edit the photographs, create the panoramas, annotate them, record data collected 

       Evaluate the itemized features, guilds and plants

• Repeat at least annually, at most 4 times a year.

• Share

How I developed this concept

I had decided I needed to inventory the natural resources on this property and had done quite a bit of internet research which turned up mostly how to estimate board feet in a forest and a few discussions of urban tree health tracking.  I posted a question in the LinkedIn Ecological Alliance group about whether it is "reasonable to do a complete inventory on a 10 acre (4 hectares) area?"  There were several thoughtful replies, all of which helped clarify my thinking about what I thought I wanted to do.

The basic problem is that a permaculture inventory implies we are going to itemize every ecosystem element as well as every relationship between and among them.  There's only about a hundred trees showing in this photo of the South West Forest. But there are thousands of ecosystem elements you can see if you  click on the image and select the original size.  So clearly there is no reasonable way for one person to conduct a complete inventory of the ecosystem elements on 10 acres of mostly woodland.

On the other hand, as Dave Jacke says in the vision part of the two volume Edible Forest Gardens,

"If you already have woodland on your property, you should first carefully inventory it. Then you can respectfully add to and subtract from the existing plant community to make your forest garden."

A forest garden is a nature-man collaboration whose success depends on man knowing the forest well enough to respect what the forest wants.   The forest is not a static thing that you pound and shape into a garden.  The forest is a community of living beings who have their own agenda, but whose community activities can be beneficial to the humans who live there.

The degree of respect given during a permaculture inventory is shown by the words recorded in the inventory.  In the image to the right, there is a fallen tree in the middle, reaching out over the wetland.  The inventory can describe this as firewood or as future mulch.  Calling it firewood respects a human use, while calling it future mulch respects its role in nature.  

After several more weeks of research and analysis, I came up with this method.  

We start with two statements 

of using my skills as a photographer and database worker a number of conclusions and directions to take.  A few hours after the first final draft of these conclusions, I came across the following quotes from a comprehensive resource on temperate climate forests: Edible Forests,

"The number of relationships among ecosystem elements is staggering, and beyond our rational capacities to comprehend."


"If you already have woodland on your property, you should first carefully inventory it. Then you can respectfully add to and subtract from the existing plant community to make your forest garden."

In other words, the basic answer to my question is that it is impossible, but do it anyway.

I count 91 trees in the photo to the right.  This view covers maybe one acre, and with 5 more acres wooded at about the same density,  there are four to five hundred trees to be enumerated.   But in between the trees, invisible at this time of year, are gooseberry patches, mayapples and lots of other vegetation which may or might not be helpful to the forest's own needs, or mine.  

I've made this photo available here in the original size which shows enough detail that most of the trees can be identified by the bark.  Click on the image, and in the lightbox select the sizes icon in the lower right corner and select original size.  

Using a photo as the basic recording mechanism for my inventory is not the same as walking up to each tree with clipboard or tablet in hand and reference books in a back pack and doing a close up and personal conversation with the tree.  But I think for the most part, I can leave the clipboard behind and record my notes when I process the photos.

Although using photography was a given for me when considering doing a ecological inventory, my primary concern was how to organize the inventory data so that extracting information from it would be reasonably easy to do.

The Southwest Forest, 2014

Looks quite open in early April. There a lot of trees, though, and finally some youngsters here and there.

By calling this a permaculture inventory, the data collection is not only about the vegetative individuals in the forest, but also their ecological relationships with all the other individuals in the forest.

A complete permaculture inventory is neither possible nor practical.

However, a comprehensive inventory is both reasonable and practical.

What is included in the inventory depends on what you want to comprehend.

What I want to comprehend at the end of each year's inventory is how well the various areas of the garden are progressing toward goals set by nature or by myself, or by some sort of compromise between me and mother nature.

My goal is for half of the property to be a "restored" native forest and the other half to be an aesthetically pleasing, edible, and healthy forest garden.

Comprehensive for my purposes is to record enough information about the members of this forest community and their interactions so that I can say I understand more of what's going on than I currently do.

 The information I need includes is basic counts of the different species in each area and  additional data about each itemized feature, guild and plant.  Over time.  I put on my database modeling hat and went through perhaps a dozen iterations before I came up with the model on the left.  This model is a high level view of what I'd like to see implemented in one of the big RDBMS, but which I will most likely implement as a set of related spreadsheets.  

A great deal of attention was paid to what kind of information I want to be extract from the data recorded.  One "report", for example, will be a season-by-season, year-by-year  sequence of photos of a particular feature, guild or plant.  Others will relate to the various kinds of ecological systems discussed in the literature.  My primary source for the kinds of data that needs to be recorded is Edible Forest Gardens

Permaculture Database Model

This design is intended to link togther an inventory of plant resources with periodic photographic records of the process. The green tables on the left define the hierarchy of locations down to the individual plant level and the related observation and view points. The pink table is all of the photos related to the forest, each tied at least to a view point. The Blue tables are the on going journal entries that give detail to the historical record. The yellow table represents internet resources to link to. The grey tables are where the data integrity and useful business rules would occur.

3. Photography can largely substitute for the kinds of detailed words needed to describe the health of a tree or of the relationships between the plants in a specific area, but only to a trained observer. With sufficiently high quality, 360 degree panoramas taken from strategically placed observation points will provide a "complete" inventory. However, the technology is not quite there to identify and count all the ecological participants shown in the photo, though there is software that will find the text and could put it in a database. Automated location determination of each item identified is possible if the camera has recorded accurate GPS data along with compass heading and altitude, etc. On the other hand, one can visually explore such a photo and identify areas and groupings to look at in more detail. My test photograph is available on gigagpan, one of the few hosts where one can share this size of image with an excellent user interface. My test was mainly to see what kind of coverage I could get and how easy it would be to annotate the image with notes about what is there. For the actual inventory images, I will need a much more precise way of identifying the compass directions, and of course will need to tweak the individual image exposures prior to stitching.

4. The level of detail needed was probably the most difficult part of the plan for the inventory. In the end I decided I needed an hierarchy of locations so that I could identify areas not too big and not too little that can be drawn on a map of the place. Each of these areas contains one or more features, a specific tree, a garden bed, something natural or something man made. In the forest and garden areas, each feature may contain one or more plant guilds -- groups of adjacent plants that can be observed and worked with as a unit. And finally each guild has at least one itemized plant, usually a tree, that is the "main" citizen of that guild. The detail information at the feature level will be more oriented toward aesthetic and human enjoyment. The detail information at the guild and individual plant levels will be oriented toward understanding the interactions and health of the part of the forest, with structures and predefined lookup tables to allow extracting meaningful reports. I have identified 35 areas and expect I'll end up with maybe a hundred features, guilds or plants.

5. The main inventory shoots done each year (season?) will be the series of 360 panoramas taken from each of the observation points. However there will be many photographs of guilds and plants taken at other times and closer in to show detail. These additional detail photos will be stored with tags or keywords indicating which observation point contains the feature, guild or plant being shown. When possible, a specific compass heading and distance from the observation point to the object will also be recorded along with the compass heading and distance to the camera for that photo. As a corollary, need to indicate in the feature, guild and plant records which is their primary observation point so that the information doesn't get fragmented, since most items will be visible from more than one observation point.

6. It might be possible to set up a series of panoramas that work about the same way those virtual tours of new homes work, allowing one to wander the paths through the forest. However, the two biggest problems for collecting that kind of imagery is first topography -- decent looking panoramas require rotating the camera around a flat axis, so that when an observation point is on a hillside, uphill is dirt, downhill is treetops. The other problems is that trees are big. It is essentially impossible to capture a whole tree, from ground to canopy, in a single image or even a vertical panorama without serious perspective problems. Even a fully articulated view camera can't deal with perspective corrections that large. Unless one can get far enough away with a clear view; but inside the forest, clear views mostly don't exist that show complete trees from top to bottom.