The Weecology group run by Morgan Ernest and Ethan White at the University of Florida is seeking a post-doctoral researcher to join our Everglades Wading Bird project. This is a long-term study monitoring wading bird breeding colonies in the Everglades, with long-term (30+ years) and large-scale (> 3000 sq km) data on nesting effort and nest success for multiple species of wading birds. We are looking for someone interested in leading research in one of two areas: 1) cross-scale drivers of wading bird breeding activity and success, 2) developing automated drone-based monitoring of nest success using aerial imagery and computer vision. Because the Everglades is undergoing one of the largest restoration efforts in the world, Weecology interacts with a wide array of state and federal agencies making this an excellent opportunity for people passionate about research with real-world applications as well as those more interested in studying general ecological processes and/or using technology to expand the scope of data collection in ecology. The postdoctoral researcher will have the opportunity to participate in fieldwork, but it is not a required activity.

At Weecology, we are committed to supporting and training a broad scientific workforce. Former Weecologists work at NGOs, big and small companies, teaching-focused colleges, and research universities. Current and former group members encompass a variety of racial and ethnic backgrounds from the U.S. and other countries, members of the LGBTQ+ community, military veterans, people with disabilities, and students who are the first generation in their family to go to college. We work hard to create a supportive lab environment for everyone in the group and expect all members of Weecology to help achieve this goal.

Postdoc funding is guaranteed for 1 year with a second year available pending no changes to funding availability. Salary is $60,000 and the position can start as early as July 1^st 2026 but the successful candidate must be able to start no later than September 28^th 2026.  Remote work is unfortunately not an option for this position, so applicants should be willing to move to Gainesville, FL.

Applicants should email Dr. Morgan Ernest (skmorgane@ufl.edu) by May 8th, 2026 with their CV including a list of three references and a cover letter detailing their research interests/experiences and which wading bird project outlined above they would be interested in working on and why.

If you’re new to the story, see the list of blog posts at the bottom, I recommend reading them in order from top to bottom. For everyone else, here’s your beginning of the episode recap:

Once a mighty force within the Portal rodent community, fall of 2023 saw the precipitous decline of the abundant Desert pocket mouse (Chaetodipus penicillatus). Thanks to deserts being addicted to extremes, it can be hard to discern blips from trends with a small number of data points but months turned into more months and the Desert pocket mouse showed no signs of recovery. Whatever forces were conspiring against the Desert pocket mouse did not seem to have the same impact on its main rival: Merriam’s kangaroo rat (Dipodomys merriami). From August 2024 to March 2025 (my last post) rain avoided the site like it was a…well, desert. We joke about the kangaroo rats being the tank of the community, but they were earning that reputation. Species after species disappeared, but they trucked on. Things looked grim and rodent abundances kissed Morgan’s Rubicon — the abysmally low abundance levels that have heralded the coming of a new regime in the past.  

Let’s pick up our story.

I wish I could tell you that the rodents at Portal have been living the Rodentia version of a life of plenty. Seeds falling from the sky like manna from heaven. We have two plant growing seasons at the site, each of which supply seeds for our rodents to eat: March/April (winter growing season) and August/September (summer growing season). Our winter plant production in 2025 was effectively nonexistent.  I don’t mean below average (which it was) I mean fields of desolation. We count plants on standardized quadrats twice a year. I’ve plotted our annual plant count for both winter and summer. You’ll notice two things: 1) life was better for plants from 1980-1994 and 2) we have been in the fields of desolation since our COVID gap (2020-2021) with the past two years being bad even by those meager standards.

Given the historically terrible plant growth for so many years, it’s no wonder that the rodents have fared poorly. After last March, rodent numbers continued to plummet. The community not only crossed Morgan’s Rubicon (red dashed line in graph below), the community explored new lows for a record setting amount of time.

Summer 2025 brought the tiniest modicum of relief. Late rains triggered a small amount of plant germination. When we surveyed the site in early September, I was certain all those newly germinated plants would die (September is late for our growing season and the tail-end of our summer rains). But the field crew in October saw seeds on plants and in December we do see a small, nearly infinitesimal uptick in rodents with our December rodent count returning upwards to touch the Rubicon. A few individuals even produced offspring, as we caught some juveniles and saw individuals in reproductive condition.

Community-wise, our once diverse rodent assemblage is down to 2 species: Merriam’s kangaroo rats and Desert pocket mouse. With so few plants and rodents to count, we trapped outside the site this summer to see how the surrounding area was faring, targeting a mixture of Portal-like habitats and Portal-different habitats (areas historically more grassy or open, cattle tanks, and shrub-thick ravines). Nowhere was brimming with rodents, but we did find other species out there, so sources of recolonization exist in the neighborhood.

As for our two remaining species. Desert pocket mouse has been below its historical monthly averages since October 2023 (red line = Jul 2023-Jun 2024). They remained this way for the entirety of 2024 (Maroon line = Jul2024-Jun 2025). And in 2025 (gold line = Jun 2025-December 2025) they only returned to the fold of the long-term average in December, a month with historically low numbers because individuals go into torpor normally.

An equivalent graph shows our kangaroo rats losing their battle with the forces of dryness. In 2023-2024 (red) kangaroo rats are doing pretty well, above average in every month. Then in 2024-2025 (maroon) they begin to slide towards the average and then cross it. By Jun 2024 they are historically low numbers for that month. Kangaroo rats remained down on the mat through the end of 2025 (gold line), but if you squint real hard you might be able to see them pushing themselves back up. Maybe.

So what does all this mean? When I started this series of posts, I assumed we were in a transition phase that we would eventually emerge from in a new and glorious state like a phoenix from the ashes. But now I wonder. Our plant data may suggest that we are 6 years into a new regime of low herbaceous plant production that our rodents have only now (possibly thanks to seed stores and the seed bank) caught up with. This of course is the problem with dynamic ecology: you often need many data points to disentangle short-term dynamics, long-term dynamics and noise, especially if there are buffering processes (like seed stores) that can generate lags in responses. So after 45+ years of data collection, the only way to answer this question is, you got it, more data collection. So stay tuned as we continue to watch the skies for rain, the ground for plants, and our traps for rodents. And hopefully I’ll be able to give more frequent updates if [waves hands at the world] let’s me.

Previous episodes:

Episode 1: Regime Shift Cometh? – Jabberwocky Ecology

Episode 2: Regime shift still cometh? – Jabberwocky Ecology

Episode 3: Regime shift cometh? Saga Continues – Jabberwocky Ecology

Episode 4: Regime Shift Status: Ongoing? – Jabberwocky Ecology

Episode 5: Regime Shift: A Winter of Grim – Jabberwocky Ecology

Our experience with machine learning for biology is that projects are not limited by models, or even data, but by the capacity for implementing workflows that can be scaled and maintained over time. The technical overhead of machine learning continues to grow. The drumbeat of new model architectures, benchmarks and ideas obscures the reality that most applications are limited by organization, data sharing and maintainability. DeepForest 2.0 is built specifically to help with these details. We have moved model sharing to Huggingface to allow easy contributions from users and sharing among research teams. We have refined prediction workflows to scale with massive new GPUs, while providing options for users with limited computing power. We focused on general biodiversity categories, ‘trees’, ‘birds’, ‘livestock’, rather than detailed taxonomic models. DeepForest 2.0 adds more support for the Detection + CropModel concept, in which two separate models are used for object detection and a second model is used for classification. We have found this to be simple, generalizable, and easier to interpret.

The next step for DeepForest is to expand the toolsets for the common workflows in airborne biodiversity monitoring. Want to annotate images with a single keypoint instead of more time-intensive boxes? Or maybe you need that extra geometric detail to get your tree boundaries? A keypoint model and a polygon model are coming soon. Have a huge dataset and don’t know which images contain biodiversity objects? An integrated active learning workflow is coming soon. At the same time, DeepForest will continue to pursue integrations with the growing libraries of machine learning tools for environmental science. Torchgeo, Pytorch Wildlife, and many other libraries are great contributions to open-science and we want to work closely with these teams to allow models and data to be shared among packages. There is much work to be done and we welcome contributors to get involved to share their experience and needs. Together we can make a better open-source ecosystem to bring AI innovations to everyone. Deepforest can only improve with user issues and contributions. We welcome issues, requests and discussions of ideas big and small to make DeepForest the best it can be.

Release Notes

There are a small number of breaking changes to DeepForest as we standardize the API. We have coalesced around Hydra for configuration management. A scalable customization tool for managing scientific experiments, hydra allows users to specify a range of configuration values and pathways.

1. Huggingface for Model-Sharing and Archiving

The explosion of machine learning for biology means there are more models than ever before. Rather than focusing on creating models ourselves, we are committed to creating a simple avenue to plug models from the community into DeepForest. Adopting functionality from the popular transformers library and model sharing using HuggingFace’s repositories, users can upload models to their own repos and get them through the new load_model structure. This works for detection or classification models and will allow us to quickly integrate new models. We welcome contributions and pull requests to list your model as available for others in the community.

As an example, we recently included support for Deformable DETR, a state-of-the-art transformer based object detection architecture. In the near future we will support different backbones for our default RetinaNet model, including pretrained foundation models like dinov3.

The use_release functions have been replaced with the general load_model:

from deepforest import main

model = main.deepforest()
model.load_model("Weecology/deepforest-tree")

2. Configuration using Hydra

Hydra is another one of those quiet but essential entries to machine learning workflows. Hydra is based on Omegaconf, a configuration manager that uses the YAML file format. Each config file inherits from a default config that we provide, which means you can easily adjust one or two parameters without having to copy and paste everything else. For example, here is how you might specify a new config to train on your own dataset, with a different batch size:

defaults:
  - config
  - _self_

batch_size: 8
num_classes: 2

train:
  csv_file: /path/to/your/annotations.csv
  root_dir: /path/to/your/image/folder

With the new command line interface, you don’t need to write a config file at all – you can set parameters as arguments to the tool (see below). This replaces the old interface where there were arguments passed to deepforest.main

model = main.deepforest(num_classes=2, label_dict = {"Oak": 1, "Maple": 2})

model = main.deepforest(config_args={"num_classes":1, "label_dict" = {"Oak": 1, "Maple": 2}})

3. Better Data Reading and Visualizations

In the theme of under-appreciated obstacles to project success, getting data into the correct format and visualizing machine learning models and their outputs is chronically undervalued. To help with data loading, a new read_file method helps read a large array of standard formats, from .csv/.json/.xml/.shp files in common formats like COCO and PascalVOC for object detection. DeepForest 2.0 adopts the supervision package from RoboFlow to help smoothly create plots of images, annotations and model outputs. Where possible we look to create connections with other open source packages and supervision will help us support a range of annotation formats, including points, polygons and videos. Check out their documentation and examples!

5. Deprecations and Name Changes

DeepForest prioritizes simple functions to complete common tasks. Users often need to convert predictions overlaid over images into geographic coordinates on the earth’s surface. We have revamped this workflow and introduced two functions image_to_geo_coordinates and geo_to_image_coordinates to replace the old boxes_to_shapefile workflow.

All 1.0 visualization functions have been deprecated including plot_predictions(). Please use plot_results().

A full list of migrations is here:
https://github.com/weecology/DeepForest/blob/main/HISTORY.md#version-200rc1-date-october-21-2025

6. Command Line Tools

Not all of our users are comfortable writing custom scripts and there are tasks like training, prediction and evaluation which require writing more-or-less the same code each time. You can now use the deepforest command to perform these actions, which also supports config overrides. You can also look at the command line script as a reference example that you could modify in your own code. For example, to train a model:

uv run deepforest train train.csv_file="/path/to/labels.csv" train.root_dir="/path/to/images"

Or to use your own config:

uv run deepforest --config-name my_config --config-path ./configs train

If your configuration has validation.csv_file and validation.root_dir, running deepforest evaluate will run our standard detection metrics on it.

Acknowledgements

We’d like to thank the World Resources Institute for providing the primary funding for recent DeepForest development as well as the National Science Foundation which funded earlier work on the software.

Previously, on Portal: the Regime Shift — In fall of 2023, one of the ruling families of Portal — the Desert Pocket Mouse (Chaetodipus penicillatus for my mammalogy friends) — fell from favor with no warning. With each passing month, hopes for a recovery faltered and then dimmed. By fall of 2024, this once thriving population had become a dismal shade of yore. For the rest of the rodents, life at Portal was still sunny, but that counts for little without rain in the desert. Few plants germinated after the fall of the Desert Pocket Mouse. Even shrubs started to die. No species was safe, and populations began to drop like characters in a George R.R. Martin book. Only kangaroo rats (Dipodomys merriami, specifically) were making a valiant stand against the forces of dryness…

That was our status as of early December 2024.

Since then, that grim summer season where ecologists counted dirt instead of plants provided few resources for our rodent friends. Their numbers have continued to dwindle.

In fact, we have now officially crossed “Morgan’s Rubicon” (thanks or blame for that name goes to Ethan). This is the (observationally) critical threshold of abundances where regime shifts tend to occur. The species that are able to hold on during this critical juncture seem to become the dominant species in the next regime. So, you may be wondering — who will be the rising rulers of Portal?

Well, it is not the Desert Pocket Mouse.

In fact, as of our most recent sampling trip, there is only 1 species (yes ONE) on any control plot (long or short term):

Given the low abundances at the site, dominating the control plots is less of a triumph for Merriam’s kangaroo rat than a testament to their grit and determination. Some plots have only a single individual.

The site was not completely bereft of diversity, however. A very small number of Desert Pocket Mice and Bailey’s Pocket Mice (a former dominant that crashed in 2010) are hiding on the kangaroo rat exclosure plots. Their absence from any of our control plots and their small numbers even on our kangaroo rat exclosures suggests they are struggling more broadly than just our study site.

So, the table has been cleared. One species has the advantage. All that is left is rebuilding. When might we see the site recover into a new regime? Not anytime soon, I’m afraid. The lands of Portal are parched. It has not rained, to speak of, since August.

This does not bode well for the spring seed harvest to sustain our rodents. Our intrepid team will count plants in early April, but we are expecting another exercise in counting dirt. Without spring plants, our rodents will have to eke out an existence on their seed stores until the summer plants (hopefully) germinate in July. Will the kangaroo rats be able to hang on to their advantage through the grim times of starvation ahead?

Stay tuned for our next episode of [cue music] “Portal: the Regime Shift”

Recap: In fall of 2023, one of our most abundant species, Chaetodipus penicillatus (Desert pocket mouse), experienced a population crash right before their seasonal torpor for the winter. Had they hibernated early or had we really lost individuals? We could only wait and see what happened in the spring. Spring came and still no Desert pocket mouse. But June and July are typically when the big population explosions occur for this species. Perhaps they would experience a rebound. Given how few individuals we had this would require dispersal from populations that had not experienced whatever had happened to ours. It wouldn’t be the first time dispersal had surprised us! (Delayed Compensation for Missing Keystone Species by Colonization) However, as of June 2024 there was no sign of a rescue coming for the Desert pocket mouse.

So, where are we now?

[Mentally insert your favorite shrugging emoji here]

TLDR; I still think the site is undergoing a major transformation. But we’re a year in and I’m no closer to being able to say whether the regime shift has happened, is happening, or what. But things are notably odd in both the rodent and plant community.

For those who want to see the details, keep reading!

Graphs all have data through October 2024. First here’s the abundance of Desert Pocket mice on our long-term control plots since January 2022. Remember, winter torpor created lulls around January every year, this has been true since the study began.

Obviously, they weren’t rescued by a major dispersal event from surrounding habitats. This also tells me that whatever happened to our population is probably a regional event. We have tons of evidence that dispersal to our site is common – species come and go and return consistently since 1977 (Regulation of diversity: maintenance of species richness in changing environments is a good place to start if you want to know more).

Previous regime shifts at the site also involved what were probably large-scale population collapses of an abundant species (Long-term community change through multiple rapid transitions in a desert rodent community). This is what caused me to start paying attention back in 2023 when our Desert pocket mouse population suddenly collapsed.

But those events were not just collapses of an abundant species – though they often seem to be the hardest hit (Catastrophic Decline of a Desert Rodent, Dipodomys spectabilis: Insights from a Long-Term Study). Our regime shifts coincide with incredibly low rodent abundances overall. Do we see this now?

The above graph shows total rodent abundance on our long-term control plots since late 1970s. The blue bars mark our four regime shifts. The red dashed line simply marks the lowest abundance in our first regime shift (mid-1980s). Every regime shift has at least touched that line. Why are our regime shifts associated with low abundances? There are a couple of possibilities. Obviously, low abundance is an inevitable outcome from a rapid population crash of an abundant species since it would take time for other species to increase through either population growth or colonization. So perhaps its just a symptom not a driver. But what if these low abundance events create stochastic or deterministic opportunities? I have watched Tadashi Fukami’s work on priority effects with interest for many years and the flip side of priority effects is incumbency – the advantage that an existing incumbent (or resident species) has over invaders or rarer species, thus resisting community change. If multiple community states are possible through priority effects, then perhaps the current one only persists because of momentum (or incumbency)? What if low abundance events create opportunities for alternative states to assemble?

But I digress.

The yellow box in the graph above highlights the current “may or may not be an event”. It has not touched the line….yet. I would be very surprised if we do not touch it sometime between now and March. Winter months typically provide our lowest abundances of the year.

The graph also shows, however, that not every low abundance event is associated with a transition event. We clearly hit low abundances in the late 1970s and in 2019 (gray box, I forgot to put one on 2019 so just imagine it there). Why do some low abundances and not others cause transitions?

If the hypothesis that crashes in abundant species create opportunities for reconfiguration is correct, then one thing that might be necessary is for the other species to be poised to take advantage. Not being as negatively impacted as the dominant would help position them for a head start in population increase (and maybe dispersal from other less impacted sites). When the 1999 regime shift occurred this was exactly what Thibaut and Brown proposed (Impact of an extreme climatic event on community assembly). Do we see this now? The data suggest that we do see differences in this current “event” in population responses.

The above graph shows relative abundance of Merriam’s kangaroo rat (another abundant species) across months. The blue curve is the long-term average relative abundance of this species in the community, showing a seasonal signal. The red line is the July 2023-June 2024 relative abundance values. You can see the jump after September when the pocket mouse crashed. The black line is July2024-present data and it just gets even higher. If all species were hit equally, total abundance would drop but relatives should stay relatively the same. [small caveat – we are dealing with small numbers of rodents at this point, so a wiggle of 1 individual can look bigger when relativized]. We are not seeing that. I can also say from my trip to the site in September that while the kangaroo rats were in reproductive conditions, very few of the pocket mice were. So, kangaroo rats are poised to reclaim their dominance of the community…maybe.

My final piece of evidence that we are going through a shift?

The shrubs are dying.

As you can see in the the picture above, many shrubs have leaves, but they were not as dense as I’ve seen in past years. It looked like I was visiting at the start of green-up instead of its peak. Large branches on many shrubs were dead or dying, like the tall skeletal branches behind the low fence. And a few shrubs had given up the ghost all together – like the tangle in the foreground. I’ve been working at the site since 1995. I have never seen anything like this.

Worried I might be an unreliable observer? Here’s our phenocam data:

First, yes, we do have troubles keeping our phenocam running. It’s Portal. It does not like electronics. Second, the big peaks you see are all summer greenness. Sadly, we only have small glimpses from 2021, but they suggest there was also a normal looking summer peak that year. My visit, on the other hand, coincided with the smallest greenness values we’ve seen since the phenocam started in 2017. Comparable, in fact, to the 2020 spring super bloom. Which sounds great, but is obviously not the same level of productivity as our summers and is also an unusual burst of productive that occurred in cooler temperatures and no shrubs leafing out. So that fact that this summer could barely compete with that is not a good omen.

So, for the intrepid few how made it this far? We are still not not in a regime shift! (love those double negatives). And honestly, this blog post has given me a whole new set of ideas for things to go look at while we all wait for the next data update….

The Weecology lab at the University of Florida has an opening for a computer vision developer to join our team. We use airborne imagery across research programs to monitor trees and wildlife and develop open source tools for the biological monitoring community. The position will contribute to research projects, Python package development, and community outreach. Areas of computer vision research include fine-grained classification, active learning and sensor fusion from data collected from UAVs, crewed aircraft, and (potentially) satellites. Our open source projects include DeepForest, the MillionTrees benchmark, as well as upcoming efforts for a wide array of challenging biodiversity applications.

Ideal candidates will work effectively with a history of responsibility and attention to detail. Independence and a willingness to take ownership over projects is key. Some experience with Python, open-source software development practices, and deep learning frameworks is preferred, but there is also ample room for skill building and mentorship within the lab composed of research scientists, software engineers and biologists. Our lab has a proven track record in developing machine learning software and datasets for ecology. Your work will be valued and used by scientists around the world contributing to biodiversity monitoring with a global reach.

We are committed to supporting and training a diverse scientific workforce. Current and former group members encompass a variety of racial and ethnic backgrounds from the U.S. and other countries, members of the LGBTQIA+ community, military veterans, and people with disabilities. We work hard to create a supportive and inclusive lab environment and expect all members of Weecology to abide by the lab code of conduct. 

Interested applicants should contact Dr. Glenda Yenni (gmyenni@ufl.edu) by November 30th. Fully remote work is possible for this position. Anticipated salary is $60,000 – $70,000 annually (depending on experience) including full benefits and there is potential flexibility in salary and work arrangements for highly qualified applicants. In your email, please include as a single attached file: 1) why you are interested in this position (300 words or less) and 2) a resume or CV including descriptions and links to prior work and/or research experience and the responsibilities, skills, and duties involved in those positions.

The student would be a member of the Weecology research group. Weecology is a partnership between the Ernest Lab, which tends to be more field and community ecology oriented and the White Lab, which tends to be more quantitatively and computationally oriented. The Weecology group supports and encourages students interested in a variety of career paths. Former weecologists are currently employed as data scientists in tech and at research centers, as scientists with NGOs, and as faculty at both research and teaching-focused institutions. We are also committed to supporting and training a diverse scientific workforce. Current and former group members encompass a variety of racial and ethnic backgrounds from the U.S. and other countries, members of the LGBTQIA+ community, military veterans, people with disabilities, and students who are the first generation in their family to go to college. We work hard to create a supportive and inclusive lab environment and expect all members of Weecology to abide by the lab code of conduct (https://www.weecology.org/lab-wiki/code-of-conduct/)

Salary starts at $29,500/year (0.5 FTE), with health insurance and a full tuition waiver. More information about the group is available through: 1) the weecology website; 2) the Portal Project website; 4) our wiki; and 5) our blog. 

Review will begin October 25th, 2024. Interested students should email Dr. Morgan Ernest (skmorgane@ufl.edu) with 1) a CV and 2) a personal statement containing information on your past research experience (what types of experiences have you had and what you enjoyed about them), your future research interests (what topics really excite you and make you want to learn more?), and what are you hoping your graduate experience will provide you? A paragraph on each is sufficient!

(Blog Post) “State-Space Vector Autoregressions in mvgam” by Nick Clark. If you model ecological time-series you probably have to deal with measurement error, relationships between time-series (e.g., interactions between species or correlated responses among regions), and non-Gaussian data (e.g., integer counts, values that can’t go below zero). This blog post does a great job of exploring how to address all of these complexities simultaneously using the mvgam R package (which we also like).

(Software) “neonutilities“. If you work with National Ecological Observatory Network (NEON) data the NEON data science team have just released a new Python package for downloading and managing their datasets. It mirrors the API of the associated R package, which we’ve been actively using for a while. Definitely recommend checking this out for Python centered research projects or as a good data source for teaching.

Ethan White’s lab at the University of Florida has an opening for a PhD student starting in Fall 2025 to study ecological forecasting of wading birds in the Everglades. This position is funded by an NSF grant to study ecological forecasting across scales using data from a combination of long-term monitoring and drone-based surveys.

The student will work as part of a team of researchers including a field crew collecting data in the Everglades, research scientists managing automated workflows for making forecasts and processing remote sensing, and Dr. Morgan Ernest, who leads the Everglades monitoring work and co-leads the forecasting research. The student will have the opportunity to develop their own research interests in ecological forecasting and community dynamics using the extensive data from the study system. The student will not be expected to lead any field work, but will have the opportunity to participate in field work and drone surveys to gain experience in the ecosystem. This position will be most rewarding for someone who is looking for a PhD experience that focuses on computational training in spatiotemporal data science and modeling, while also providing opportunities for field work and engagement with remote sensing.

The lab is part of the weecology research group. Weecology is a partnership between Morgan Ernest’s lab, with expertise in field ecology and community dynamics, and Ethan White’s lab, with expertise in environmental data science. The weecology group supports and encourages students interested in a variety of career paths. Former weecologists are currently employed as data scientists in tech companies and academic research centers, as scientists with NGOs, and as faculty at both research and teaching-focused institutions. We are committed to supporting and training a diverse scientific workforce. Current and former group members encompass a variety of racial and ethnic backgrounds from the U.S. and other countries, members of the LGBTQIA+ community, military veterans, people with disabilities, and students who are the first generation in their family to go to college. We work hard to create a supportive and inclusive lab environment grounded in the lab code of conduct.

Salary starts at $29,500/year (0.5 FTE), with full tuition waivers and health insurance. More information about the group is available through: 1) the weecology website; 2) the Everglades project website; 3) our GitHub organization; 4) our wiki; and 5) this blog. 

Preliminary review will begin October 1st, 2024. Interested students should email Dr. Ethan White (ethanwhite@ufl.edu) with 1) a CV and 2) a personal statement containing information on your past research experience (what types of experiences have you had and what you enjoyed about them), your future research interests (what topics really excite you and make you want to learn more?), and what you are hoping your graduate experience will provide you? A paragraph on each is sufficient!

Lots of things we like and want to share don’t justify an entire blog post (or at least the time it takes to write one), so we’re going to start some semi-regular posts title “Things we(ecology) like(s)” where we share a couple of sentence and an associated link to one or more things we’re excited about that you might enjoy learning more about. If you follow Dynamic Ecology you can think of it as our version of their Friday Links posts, but likely a bit shorter on average.

(Paper) “The Everglades vulnerability analysis: Linking ecological models to support ecosystem restoration” led by Laura D’Acunto & Stephanie Romañach from the U.S. Geological Survey. This is a really interesting paper developing an integrated decision support tool that makes predictions for a range of ecological responses of interest to the management of the ecosystem including vegetation type, peat accumulation, alligator nesting potential, and wading bird colony counts. We like how it integrates predictions for all of these groups using modular Bayesian networks and the approach of making prediction more tractable by discretizing responses.

(Blog/Newsletter) “hrbrmstr’s Daily Drop“. I can’t recommend this newsletter strongly enough for keeping up with the latest in useful data science adjacent tech. I only read one newsletter everyday – and that’s @hrbrmstr‘s Daily Drop.