When you choose to publish with PLOS, your research makes an impact. Make your work accessible to all, without restrictions, and accelerate scientific discovery with options like preprints and published peer review that make your work more Open.

PLOS BLOGS The Official PLOS Blog

Not Seeing the Grass for the Trees

I appreciate repetition. My favorite class in high school was AP Chemistry, but I think I owe most of my AP success to the previous year’s slog through regular Chemistry. By the time I took AP Chem, all the basics were finally settled in my mind and I could hit the higher-level concepts that I’d only whiffed at the year before. The second time I ran a marathon I had so much more fun — and so much more left in the tank for that last, terrible 10K. The second time I read a paper from a new-to-me sub-discipline, or with specific, sophisticated statistical methods, I get a similar boost of adrenaline and understanding. These déjà vu methods* are often serendipitous — it’s hard to consciously search for them — but they are so rewarding when I stumble into them.


Déjà vu methods struck PLoS ONE this winter in the form of land cover change captured in Landsat images. The Landsat program has been capturing satellite images of Earth since the 1970s, and researchers can compare decades of Landsat images to tell a time-lapse story of changing land cover at a high resolution. Two great papers used this method to explore trends in forest cover on opposite sides of the globe: New England in the United States and the Western Ghats in India.


As I read Thompson et al.’s ‘Forest loss in New England: A projection of recent trends’ I could imagine the last twenty years of land cover change in my mind’s eye before even glancing at the figures. I’ve lived in New England nearly my entire life; I went to grad school in one of the three case study sub-regions, and worked college summers in another. The third case study sub-region is a long stretch of coastal Maine that I drove through every field season on my way to Acadia National Park. I’ve collaborated on remote sensing work, but it’s not my wheelhouse, so reading Thompson’s paper allowed me to enter this world in a really intuitive way because the results and projections already felt familiar to me.


And then I read ‘Not seeing the grass for the trees: Timber plantations and agriculture shrink tropical montane grassland by two-thirds over four decades in the Palani Hills, a Western Ghats Sky Island.’ Here, Dr. Arasumani and a team of academic and non-academic researchers used four decades of Landsat images to quantify patterns of land cover change in grasslands, forests, plantations and agriculture in a region of southern India called the Palani Hills. This is a landscape that I could not imagine — fortunately Arasumani’s team collaborated with photographer Prasenjeet Yadav who created an amazing 4-minute film. This video is a fantastic introduction to the ecosystem with beautiful footage of the shola grasslands and shola forests.



‘Not seeing the grass for the trees’ is a response to the local perception that timber plantations have replaced shola forests in the sky island of the Palani Hills. Local conservation policies center on restoring shola forest trees, with little focus on the shola grasslands. So, a group of scientists began using Landsat to challenge the current conservation view. As Dr. Milind Bunyan and Dr. Robin Vijayan write,


“The popular discourse that timber plantations are invading shola forests runs deep and wide, but there are exceptions to this observation. In the state of Kerala for instance, there is growing appreciation that it is the grasslands that have been lost to plantations and not the forests. The state that holds a majority of this ecosystem both in original and modified states (viz. Tamil Nadu) however, largely believes that plantations have invaded forests (although there are individuals in the state forest department who now recognize the loss of grasslands).”

These sky islands are home to this unique Shola forest-grassland complex where the forests are in the valley and the grasslands are on the hills. photographer: Prasenjeet Yadav

Coauthor Robert Stewart and his late wife Tanya Balcar had been working in the Palani Hills since the 1980s: their Vattakanal Conservation Trust focused on forest and grassland conservation and they were among the first to notice that the grasslands were disappearing. The story of how Tanya Balcar’s observations snowballed into this paper is a lovely peek behind the curtain of conservation research: the collaborations, the shoestring budget, the surprises, and the great food all ring true to my experiences working with NGOs and government agencies in New England. Bunyan and Vijayan gave me a long version to edit down, but I love the details too much.

“[Tanya and Robert] convinced some of us who were working on different projects in this landscape including Ian Lockwood, a two-generation resident of Palani Hills, and a friend of Tanya & Bob. Using his skills as a geographer, Ian conducted a preliminary GIS analysis, which revealed the dramatic changes that had occurred during his lifetime; he then published these results on his blog. This caught our attention when we realized that much of the change in the landscape had occurred very recently, and providentially within the period of LANDSAT imageries.

Anil and Sunayana Choudhary from INTACH Kodaikanal (listed in the acknowledgements) were the people who really made the project happen. They generated INTACH funding for the project to conduct fieldwork, and to hire a technician to do the GIS and ground-truthing. As with most research projects however, we did not stick to the script and ended up hiring two technicians (one for the lab and the other for the field), despite uncertainties at the time on how we would support both of them. Of these, Danish Khan came with a tremendous wanderlust and was therefore, the natural choice for our field component, and Arasumani M., who graduated at top of his class, was the lab person conducting the GIS analyses.

The only thread that binds all of us is a desire to work in the landscape, albeit on varied aspects, and understand and document the threats and changes in this landscape, which required a baseline that we could use for future studies. This was an extremely frugal study, and most of us contributed significant amounts of time (and in some instances, money) to the project in different ways. With different roles on the project, we found working together relatively easy and complementary. A lot of our work also involved working different physical locations (including putting these responses to your questions together), and used cloud-platforms like Google Docs. We’re also proud to say that our meetings were almost like large family gatherings, full of great food (supplied generously by the Choudharies), and travelling through the landscape.”


Matrix of landscape modification. photographer: Prasenjeet Yadav

Through Landsat images and ground-truthing, this team found that shola grasslands — the dominant cover type forty years ago — had been invaded by agriculture and plantations. Agriculture and plantations overran shola grasslands with different spatial patterns of replacement and degradation: agriculture takes over in “large, compact, and spatially aggregated patches” while plantations puncture the landscape with small, irregular-shaped patches as invasive plantation species spread into the ecosystem. This analysis also found that only half of the existing grasslands are currently included in the Kodaikanal Wildlife Sanctuary; they identified eight additional grasslands along cliff edges or bordering abandoned agricultural areas to include in this sanctuary. Finally, the authors conclude with four specific conservation recommendations: (1) identify and conserve core grasslands (2) check invasion in sparsely invaded grasslands (3) review indiscriminate removal of mature plantations (4) contain agriculture. I asked Bunyan and Vijayan how these recommendations have been received by the Kodaikanal Wildlife Sanctuary and the local communities. They write “In addition to our town hall meetings, we have had several interactions with forest department staff of the KWS to disseminate our conservation recommendations. We opine that the Forest Department is positive, and we hope to work with them to be able to achieve the goals stated in the paper. The publicity that this article has generated, which has been covered in the national media and now internationally, will go a long way in promoting these recommendations.”


I wish good luck to them as they continue this important work! And thank you for enriching my winter with beautiful images of the shola grassland!

The Nilgiri Pipit, a threatened grassland endemic bird. photographer: Prasenjeet Yadav


Arasumani M, Khan D, Das A, Lockwood I, Stewart R, Kiran RA, et al. (2018) Not seeing the grass for the trees: Timber plantations and agriculture shrink tropical montane grassland by two-thirds over four decades in the Palani Hills, a Western Ghats Sky Island. PLoS ONE 13(1): e0190003. https://doi.org/10.1371/journal. pone.0190003


Thompson JR, Plisinski JS, Olofsson P, Holden CE, Duveneck MJ (2017) Forest loss in New England: A projection of recent trends. PLoS ONE 12(12): e0189636. https://doi.org/10.1371/ journal.pone.0189636





* To expand on déjà vu methods, I present the thylacine. Last November I read a preprint of a paper on thylacine extinction; I don’t actually know anything about thylacines, but my friend Kevin Burgio was a coauthor, I had studied abroad in Australia when I was in college, and I thought it sounded cool. This thylacine paper introduced me to Bayesian Extinction Estimators and less than a month later, my PhD advisor published ‘A statistical estimator for determining the limits of contemporary and historic phenology’ — a paper that repurposed Bayesian extinction estimators for historical and herbaria-based phenology data. Reading the thylacine paper serendipitously primed me to fully understand this methodological approach for my own field (plant phenology). I’m not the brightest crayon in the box, but if I just keep reading déjà vu methods, I’ll make it to razzmatazz.


Leave a Reply

Your email address will not be published. Required fields are marked *

Add your ORCID here. (e.g. 0000-0002-7299-680X)

Back to top