Fothergilla gardenii - Dwarf fothergilla

This is one of several plantings that went along with the construction of the Environmental Center here at Pace - Pleasantville. In fact, this picture is of a shrub just outside the door to my building.

When I first saw it, I thought it might be some sort of alder, based on the leaf shape. I looked through all the books I had available to me, but didn't find anything convincing. So I asked our local naturalist and he said "Forthergilla". From there, I went to the internet - first stop, Wikipedia. The genus Forthergilla has the common name Witch Alder. According to the Wikipedia entry, there are only two or three species in the genus. The family is Hamamelidaceae, the witch-hazel family. So while alder is in the common name, this species isn't in the same family as the genus Alnus (Betulacea - the birches). It made me pretty happy that at least I wasn't the only person to think of this resemblance.

So why couldn't I find this plant in any of my books? Well, most all of my guides are northeast based. And my favorite website for identification, GoBotany, is also northeast based. According to information from the Missouri Botanical Garden entry for F. gardenii, this species is native to the southeast US. This makes me wonder if I should start expanding my collection of guides.

Linden viburnum - Viburnum dilatatum

I went for a nice walk with my cousin in Rockefeller State Park Preserve yesterday evening. He identified a number of the plants there for me. This one is of particular interest, because it is a non-native reported as becoming more widespread in natural areas.

Viburnum dilatatum

Honestly, I don't think it would have stood out to me as anything out-of-the-ordinary. Very little is known about the potential ecological impacts of its spread, but it is assumed that it can crowd out natives, particularly along forest edges. It is native to east Asia and commonly cultivated/planted in gardens throughout the mid-Atlantic USA.

Some distinguish characteristics:

• Gleason and Cronquist note that it lacks stipules and it's leaves are hairy on both sides. I don't recall noticing this in the field, but I also wasn't looking carefully. (I tried to zoom in to the pic, but no luck.)
• The fruits are apparently quite showy. I'll keep my eye out for this in the future.
• The UCONN Plant Database also had a good entry for this species, which pointed out features like pubescents on both surfaces of the leaves, but more densely so near veins, and the presence of bright orange lenticels. 

I'll keep an eye out for these features next time I'm botanizing in the park!

Nemesia

I like spending time in botanic gardens. For the purposes of becoming better at plant identification, it's nice to be able to look at a plant and have a label nearby that tells me 'the answer'. However, in most gardens, many of the labeled specimens are exotics or hybrids. Nevertheless, looking at the plants and studying their characteristics, then connecting them to the genera listed, still seems like a good way for me to learn more.

Two weeks ago we made our first trip of the season to our local botanic garden, Old Westbury Gardens. Being so close, we became members here last year, and find that it's a great place to bring our daughter for a few hours of running around outside. These little plants caught my eye, and I was pleasantly surprised to see that they're native to South Africa. Well, to be clear, as someone who studies invasion biology, I didn't think that was so great, but as someone who's spent sometime studying the flora of the Cape Region of South Africa, seeing these guys made me feel a bit nostalgic.

Nemisia hybrid

Their were a number of different varieties, all equally pretty I think.

Nemesia fruiticans

This last one was particularly stunning.

Nemesia hybrid

While writing this post, I looked up Nemesia in the book Field Guide to Fynbos. I was happy to see the genus in there, but not surprised that N. fruiticans (they only one of these three with a species name) was not listed. This field guide is not exhaustive. But the generic description sounds pretty right - leaves opposite, variously toothed, flowers solitary, combinations of white, yellow, orange, pink, or blue, calyx 5-lobed, corolla 5 lobed and strongly 2-lipped. There are about 60 different species of these little guys, about 25 in the fynbos. 

Magnolia 'Elizabeth'

Spring is definitely coming in the NY metro area. This past Sunday I spent a few hours walking around the Planting Field Arboretum with my daughter. It was a great day, just a bit cool (around 50F), and oddly, a few hours before we would get an inch of snow. We saw many plants blooming, but the one I want to talk about in this post is the Magnolia tree we saw.

I love the site of flowering trees in the early Spring!

To be honest, I'm still not 100% certain this is a Magnolia, and I haven't been able to get it to species. The first place I looked was the Trees of New York: Native and Naturalized by Donald Leopold. But the Magnolia species described in there didn't quite fit what I was seeing. In particular, the bud on this tree is fairly big and hairy, which didn't match the buds described in this book.

Bud, not so focused. And bark of relatively young twig.

I tried looking through a few of my other books, but eventually went to the internet and started Google searching terms like "fuzzy flower bud early spring" and what not. Ultimately, the images and descriptions I found that best fit my observations were for Magnolia 'Elizabeth'. This is a cross between M. acuminata (Cucumber tree) and M. denudata (Yulan magnolia), so a native and a exotic, respectively. It was neat to read on the Missouri Botanical Garden site that this variety is patented by the Brooklyn Botanical Garden. All things considered, it makes a lot of sense that this tree would be planted at an estate along the north shore of Long Island.

My daughter, enjoying the early spring day outside, while dad takes pictures of plants.

Japanese barberry - Berberis thunbergii

My wife, daughter, and I went for a nice walk the Friday after Thanks Giving at Fahnestock State Park. Along the way I noticed a fairly large infestation of Japanese barberry. This is a plant that I'm fairly confident in my ability to identify in the field, but I wanted to use this opportunity to learn more about its characteristics.

I usually ID this based on the red fruits, straight spines/ thorns, and the surrounding environment and plant community. However, in the past, I've not bothered thinking about the difference between Common barberry (Berberis vulgaris), which is native to Europe, and Japanese barberry (B. thunbergii). One of the key differences is that Common barberry has sharply toothed leaf margins, while Japanese barberry has 'entire leaf margins' (no teeth). It's late Autumn in NY. Looking at this picture, it's clear that leaf characteristics are not going to help me.

Another trait difference is that Common barberry as 3-pronged spines. I didn't recall seeing this in the field. When I zoom into this picture, it becomes pretty blurry, but I'm pretty sure that each of the spines is by itself. So I'm going with Japanese barberry as my ID.

Both Japanese and Common barberry are in the Berberidaceae family (i.e., the barberry family). There's some nice information on this family both in the above referenced Wikipeadia page and in eFlora. Seems to be a fairly species rich family (ca. 650 species), but relatively few here in northeast North America (based on Flora Novea Angliae).

Sources used:
previous knowledge
Flora Novea Angliae
USFS Weed of the Week information sheet (Common barberry)

American crab apple - Malus coronaria

There are a lot of plants in my own backyard that I'd like to work on IDing here. This is the first. American crab apple or wild crab apple or sweet crab apple ... let's just go with Malus coronaria. This tree caught my eye because the fruit have turned a beautiful yellow / gold color this fall (and presumably every fall). I'm not fully convinced it's M. coronaria though. From what I saw, it lacks the thorn-like structures at the end of the twigs. That might be just based on the branch I pulled off the tree, or it might be some sort of cultivar. Or perhaps I just got it wrong.

Sources used:
GoBotany
Trees of New York State Native and Naturalized
Trees of Eastern North America (Princeton Guides)

Changing the focus of this blog

I started this long neglected blog to write down ideas I had as worked my way through getting my PhD in Ecology and Evolution. I've since gotten said degree, finished my postdoc, and started as an assistant professor at Pace University in New York. I've also started a different blog (of sorts) using Github pages. 

So where does that leave this site? I had thought of continuing to neglect it for a while, until I finally just closed it all together. But today, I had another idea. You see, while I fancy myself a plant ecologist, I'm not especially good at knowing the names of many plants. I've wanted to get better at this for a long time, and perhaps some day I will. So I'm going to use this space to post pictures of plants I try to ID. 

Did I say local field work? Going international!

One of my earliest posts here was about why I enjoyed local field work (and some of the pitfalls of it too). Well, I'm writing this post from Calvinia, South Africa, a location that is definitely not local for me. I am here as part of my postdoc work, collecting data on plant functional traits  in one of the most diverse floras in the world with a group of about eight people. As the postdoc on this project I had to do quite a bit of leg work before getting here related to examining existing data from the region, putting together species lists, helping to determine optimal sampling locations, etc. Fortunately, because the research team I'm with has carried out several similar studies in various parts of South Africa, most of the challenging logistics were worked out by others with much more experience. To that end, all I really need to do is learn how to drive a standard, on the left side of the road. But putting together a species observation list and sampling location plan was definitely a new experience for me. When I started a few months ago, I had imagined that I would have to have things planned out roughly to the day, as our time is limited and the area we are covering is vast. I quickly learned that what we're after is more of a sketch of a plan. What that means is that most every night I'll sit down with the botanist in our team and help him decide on the best spots to sample the next day. And our species list is getting updated daily - checking off what we've observed, dropping some species, adding others. Really, the key is to be flexible. However, this flexibility requires preparation. It's best if I can load up the vegetation survey data a colleague collected several years ago in a matter of minutes, and the hours of examining these data before hand has definitely helped me to navigate it on the fly.

Planning considerations aside, scientifically the most important part of this trip is for me to just get out into these incredible plant communities and to experience the field and lab work involved in our project. I've been spending the past year organizing, cleaning, and analyzing data collected in previous years without fully understanding what the system looks like. I've read a lot about it, but have never seen it with my own eyes. Now I can better understand some of the ambiguities in our datasets. For example, I've now held a specimen in my hand and had the thought "what the heck is the leaf on this thing?" This experience is absolutely invaluable for me while I'm wrapping up some of the analyses I've started, and moving forward with new ones. I'm hoping to have a few posts on some of my insights (or causes of confusion!) during the rest of this trip.

On a personal level, this trip is also my longest international trip, and I gave a lot of thought into packing, and trying to be minimal about what I brought (even if I did most of my packing the day before I left). I didn't want to bring anything I wouldn't use (except my small first aid kit. I'm ok carrying that around without using it). So far it's working out well. I promise more biologically relevant pictures in the future, but for now, here's a pic of what I thought I would need for multiple weeks of field work. (Note - Nessy, the green monster, didn't come. The shoes are my wife's (though I'm told they are comfortable). And yes, one of my carry-on items did end up being the Leaf Area Meter. I'll try to sucker one of the other group members into dragging it home!)

Reflection

Today, a bit of a story ...

Kinsman Pond from ledges of North Kinsman

I stood near the edge of the clearing with the two scientists, straining my ears to hear the call of a Bicknell’s Thrush. They played the recording, and we listened. As I recall, it was a relatively cool summer evening, though it was often cool in the evening at Kinsman Pond campsite. It was 2002. I had just finished my third year of college, and I was working as a backcountry campsite caretaker for the Appalachian Mountain Club. I had been dreaming of having this job for at least two years at that point, and a few cold, rainy days aside, I was enjoying every moment of it. It was a stark contrast from the previous summer, when I worked at Brookhaven National Labs as a research assistant for one of my professors. That too was a great experience and important for my development as a scientist. But living in the woods, spending my days doing trail work, sleeping in a canvas tent, dealing with black flies ... that was the life!

On this particular evening, two scientists from the Vermont Institute of Natural Science came to my site, and were planning to spend the night so they could survey for Bicknell's Thrush both that evening and very early the next morning. We chatted for a while, and they told me that they were using known locations of this bird, along with environmental information from those locations, to build a model to predict where it might be observed. I had no knowledge of species distribution modeling, and honestly, probably only a vague notion of what a "species niche" was. What I did know was that I as a physics major, I knew what the word model meant. I knew that any such model would likely be constructed using some computer programming language. I naively thought you would probably code that up in C or C++, because hey, isn't that what most scientists were using? But most importantly, I thought "I could probably learn to do that". 

I've told this story many times over the last 7+ years, in part as a way of explaining to people how I went from physics to ecology. To be honest, the path from physics to ecology was not direct. I finished my degree, spent two more years working as a caretaker, and then spent two years working in a neuroscience research group. But this experience stuck with me, and eventually I got to "build ecological models". I was reflecting on this again recently and I finally went and looked to see if any publications resulted from this field work these scientists had been doing. Here's what I found:

Lambert et al. 2005. A practical model of Bicknell's Thrush distribution in the northeaster United States. The Wilson Bulletin 117(1): 1-12.

And here was the line I was looking for - "during 2000–2002 ... we recorded the presence or presumed absence of Bicknell’s Thrush on 130 mountains first sampled by Atwood et al. (1996) (New York: n = 30, Vermont: n = 56, New Hampshire: n = 26, Maine: n = 18)." I'm assuming that one of the 26 New Hampshire locations was Kinsman Pond.

Perhaps the best part of finding this paper is that since that meeting in 2002, I have acquired the knowledge necessary to understand what these researchers were working on. The species distribution model that they constructed is pretty uncommon compared to today's methods. It appears to be a mixture of quantile regression, which is reminiscent of some of the first distribution models used (e.g., Bioclim), and basic pattern observation. The model options for presence-absence data are quite numerous now, and it would be interesting to see how different the distribution predictions are using more modern techniques. 

It's nice to look back on this story and think "yeah, I did learn how to do that."

Stratified random sampling in R using dplyr

Stratified random sampling with dplr

Matthew E. Aiello-Lammens

July 10, 2014

Setup

Let’s say I have a number of sample units for which I have observed some characteristic(s) at two time-points. In my specific case, I have species abundance data for 120 plots in 1992 and 2011. Using these data, I calculated the species turn-over between the two time points for each plot. I then shuffled the 2011 plots, leading to random pairing of plots between the two time points, and recalculated the turn-over.
There are many cases in which we may want to do something similar to this, and many non-parametric randomization methods use a similar setup. The particular problem I faced is that the plots were stratified into broad vegetation types, Fynbos, Thicket, and Grassland. When shuffling the 2011 plots, I wanted to shuffle plots only within their vegetation type. I thought up of a number of complicated ways to write a function to do this, and even started coding one up. Then I thought about how I could use dplyr to carry out stratified random sampling. Here’s an example of how it works.

Make a data set

Here is a sample data set including 20 plots (p1, …, p20), randomly assigned into one of three categories. I’ve printed out the data set, since it’s small.

## Load dplyr
require( dplyr )

## Loading required package: dplyr
## 
## Attaching package: 'dplyr'
## 
## The following objects are masked from 'package:stats':
## 
##     filter, lag
## 
## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union

## Make data.frame
df <- data.frame( plot = paste( "p", 1:20, sep = "" ),
                  category = sample( x = letters[1:3], size = 20, replace = TRUE ),
                  stringsAsFactors = FALSE )

## Print data.frame, arranged by category
print( arrange( df, category ) )

##    plot category
## 1    p3        a
## 2    p8        a
## 3   p12        a
## 4    p1        b
## 5    p2        b
## 6    p4        b
## 7    p5        b
## 8    p9        b
## 9   p15        b
## 10  p17        b
## 11  p19        b
## 12  p20        b
## 13   p6        c
## 14   p7        c
## 15  p10        c
## 16  p11        c
## 17  p13        c
## 18  p14        c
## 19  p16        c
## 20  p18        c

Simple shuffle

Shuffling plots, disregarding their category classification, is easy - just use sample. Below I’ve printed out the shuffled paired-plots.

## Shuffle plots
plots_shuffled <- sample( df$plot )

## Print plots and plots_shuffled together
print( cbind( df$plot, plots_shuffled ) )

##             plots_shuffled
##  [1,] "p1"  "p5"          
##  [2,] "p2"  "p13"         
##  [3,] "p3"  "p16"         
##  [4,] "p4"  "p14"         
##  [5,] "p5"  "p1"          
##  [6,] "p6"  "p3"          
##  [7,] "p7"  "p20"         
##  [8,] "p8"  "p2"          
##  [9,] "p9"  "p11"         
## [10,] "p10" "p19"         
## [11,] "p11" "p10"         
## [12,] "p12" "p17"         
## [13,] "p13" "p9"          
## [14,] "p14" "p7"          
## [15,] "p15" "p6"          
## [16,] "p16" "p8"          
## [17,] "p17" "p15"         
## [18,] "p18" "p4"          
## [19,] "p19" "p12"         
## [20,] "p20" "p18"

Stratified random sampling (shuffling)

But what if we want to account for the category classification? Here’s how I used dplyr to perform stratified random sampling.

## Use dplyr group_by and mutate to randomly sample within category
df <- 
  group_by( df, category ) %.%
  mutate( strat_rsamp = sample( plot ) )

print( arrange( df, category ) )

## Source: local data frame [20 x 3]
## Groups: category
## 
##    plot category strat_rsamp
## 1    p3        a         p12
## 2    p8        a          p3
## 3   p12        a          p8
## 4    p1        b          p5
## 5    p2        b         p20
## 6    p4        b          p9
## 7    p5        b         p15
## 8    p9        b         p17
## 9   p15        b          p1
## 10  p17        b          p2
## 11  p19        b          p4
## 12  p20        b         p19
## 13   p6        c         p11
## 14   p7        c         p14
## 15  p10        c          p7
## 16  p11        c         p10
## 17  p13        c         p16
## 18  p14        c         p18
## 19  p16        c         p13
## 20  p18        c          p6

We could also return just a vector of the shuffled samples, without the data.frame. Convenient, but not very pretty code-wise

( group_by( df, category ) %.%
    mutate( strat_rsamp = sample( plot ) ) )$strat_rsamp

##  [1] "p5"  "p20" "p8"  "p9"  "p15" "p16" "p11" "p3"  "p19" "p18" "p14"
## [12] "p12" "p13" "p10" "p4"  "p6"  "p1"  "p7"  "p17" "p2"

Conclusion

There you have it - stratified random sampling. There may be an even easier way to do this (perhaps I missed a function or didn’t dive into sample enough?), but this seems pretty easy to me. Thanks dplyr!