El viernes 13 de marzo del 2026 hicimos algo que nos gusta mucho desde la Red Colombiana de Fototrampeo y la Sociedad Colombiana de Mastozoología, y es abrir las puertas para compartir nuestra experiencia en un curso de Monitoreo de Biodiversidad con Trampas Cámara y Modelos de Ocupación con R, hubo una sesión teórica y dos practicas con mucho código, datos, modelos y mapas apareciendo en pantalla en tiempo real. Nos acompañaron personas de multiples organizaciones provinientes de Bogotá, Bucaramanga y Pereira.

Lo que hemos hecho

Laín Pardo sentó las bases mostrando el aspecto histórico y la importancia de la pregunta de investigación en el fototrampeo. Mauricio Vela nos contó sobre el paso siguiente al modelado de la ocupación y es entender el movimiento de los animales y la importancia del rango de hogar para diseñar un buen muestreo en los modelos de ocupación. Angélica Díaz ilustró el uso de wildlife insights y como esta herramienta ahorra tiempo valioso y facilita la colaboración. Luego vino una inmersión al aspecto histórico de los modelos de ocupación y la importancia de los muestreos repetidos para solucionar el problema de la detección imperfecta.

El segundo y tercer día fueron totalmente en la práctica con código en R.

La dinámica fue completamente en vivo: con código real, datos reales de la Cordillera Central de Colombia, y modelos y mapas construyéndose en tiempo real. Eso es lo que más valoramos en este tipo de experiencias!

📅 Ver el programa completo

El material del segundo dia esta disponible

Si no pudiste asistir al curso en vivo, o si estuviste y quieres repasar un poco, acá está el recurso del segundo dia:

👉🏻 Codigo y Tutorial

¿Como seguir o profundizar más?

Si quieres repasar, aprender más o profundizar tus habilidades con R

👉🏻 Acá un tutorial

Si quieres profundizar en la importancia de las simulaciones y el concepto ecológico y matemático del modelo de ocupación mas básico.

👉🏻 Acá otro tutorial

La sede del curso

Este curso fue posible gracias a la colaboración de múltiples organizaciones y al trabajo de Yeimy Castillo y Mauro quienes desde el Parque Jaime Duque empujaron para que todo saliera de la mejor forma.

Foto del grupo en la mano del Parque Jaime Duque.

Por primera vez tuve la oportunidad de dar clase en el Taj Mahal… jejejeje…

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

I woke up this morning to an email I wasn’t quite expecting — mammalcol, my R package for exploring the mammals of Colombia, had been accepted to CRAN. I sat there for a moment, coffee in hand, feeling a mix of relief and quiet pride. It’s been quite a journey to get here.

It started with a messy camera trap dataset

Like most things I build, mammalcol wasn’t planned. It grew out of necessity. I was working through a large camera trap dataset, trying to put proper species names and details to hundreds of detections. I kept writing the same kinds of helper functions over and over — little snippets to look up species info, check taxonomy, pull distributions. Eventually those snippets became functions, and the functions found a home in a personal package that lived only on my computer, tangled up with a bunch of other stuff that was probably only useful to me.

When a manuscript started taking shape from that analysis, I hit a wall. My analysis code loaded my personal package, which meant no one else could realistically reproduce my work. I needed a better solution.

Pulling it apart and making it shareable

The answer came from looking at what others had done in similar situations. I should separate out the useful bits, clean them up, and publish them properly. So that’s what I did — I pulled the relevant functions into their own standalone package, invited some other people who care about Colombian mammals and R to join in, and mammalcol started to take shape as something meant for the world, not just for me.

That process of going from “functions on my computer” to “package on CRAN” was honestly more painful than I imagined. I’d heard it was tricky, but I underestimated just how much attention to detail CRAN submission requires. Several rounds of submissions. Notes coming back about small things — a period missing here, a line too long there, a word in the description that CRAN didn’t like. Each time I’d fix the notes and resubmit, half expecting another round of feedback.

A few things that helped keep me sane through the process:

• usethis — an incredible package that automates so much of the scaffolding involved in building and maintaining an R package. I leaned on it constantly.
• The R Packages book by Hadley Wickham — I followed it closely to get the first draft off the ground. A genuinely great guideline.
• testthat — writing tests felt like extra work at first, but it saved me from breaking things more times than I can count.

And then, the email

After all of that, the acceptance email felt almost anticlimactic in the best possible way. A small, quiet moment at the start of an ordinary morning. But I know how much effort went into it — from me and from the collaborators who joined the project and helped shape it into something more than I could have built alone.

The package isn’t flashy. It does a humble but useful job: giving researchers quick, reliable access to the list of mammal species of Colombia, with tools to search, map, and validate occurrence data. Colombia is one of the world leaders in mammal diversity, and I hope mammalcol makes it a little easier for people to explore that.

A manuscript is still in the works, which is what started all of this in the first place. But for now, I’m just glad the package is out there.

If you want to try it:

install.packages("mammalcol")

And if you find a bug, have a suggestion, or spot a species in a new departamento — I’d love to hear from you. The GitHub repo is the best place to reach out.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

BirdNET-Go includes a powerful feature called Sound Level Monitoring. This is crucial for understanding the acoustic context of the environment, which helps in filtering unwanted noise and provides invaluable sounscape data for researchers.

The system registers sound levels in 1/3 octave bands, following the ISO 266 standard, providing a detailed frequency breakdown of the soundscape. The data is recorded as frequently as every ten seconds, offering continuous, advanced audio monitoring that gives a far richer picture of the overall soundscape than just a list of bird names. But I selected to get data each 30 seconds to have just two sound levels samples per minute.

Getting the data

In the past post, I explain more details about the sound levels and my custom solution.

At the end I have a CSV file that grows as fast as a Mega per day.

Plotting the data

To plot the sound levels I wrote the following R code:

library(readr)
library(tidyverse)
library(ggExtra)
library(viridis)
library(hms)

soundlevel_data <- read_csv("C:/Users/usuario/Downloads/soundlevel_data.csv", col_types = cols(timestamp = col_character()))

soundlevel_data$datetime <- ymd_hms(soundlevel_data$timestamp)
soundlevel_data$dia <- date(soundlevel_data$timestamp)
# Using lubridate to extract individual components and make hour
time_string <- as_hms(paste(hour(soundlevel_data$datetime),
                          minute(soundlevel_data$datetime),
                          second(soundlevel_data$datetime), sep = ":"))

soundlevel_data |> #filter(day(datetime)==c(22,23)) |> # filter one day
 ggplot(aes(x = time_string, y = frequency, fill = noise)) +
      geom_tile(height=.11) + # trick to remove white line
      # scale_fill_gradient(low = "yellow", high = "red") +
      scale_fill_viridis(option = "H") +
      scale_y_log10() +
      scale_x_time(date_breaks = "1 hour", date_labels = "%H")+
      facet_grid(rows = vars(dia)) +
      xlab("Hour") +
      ylab("Frequency (Hz)") +
        labs(fill = "Noise (dB)") +
   removeGrid() # + #ggExtra 
  # theme_classic() 

Here the graph

I tried to match the time with some acoustic events…

Why the sound levels are in negative scale?

In my BirdNET-GO data, the reference (0 dB) is set to the maximum possible level of the microphone (without distortion) that I use to capture the sound. In my case I am using a cheap MAONO USB Lavalier Microphone. It says is capable of recording at 192KHZ/24BIT and has an audio sensitivity of 30 Decibels. So -80 dB in the graph mean it is very quiet and 0 is very noisy, not real decibels. Perhaps I need to do some type of calibration here.

My readings range from about -45 dB to -89 dB, this suggests a relatively quiet environment (birds, ambient noise, and low noise). The higher frequencies (10-20 kHz) are much quieter (-80 to -89), which is normal for rural or natural soundscapes.

Why this scale is useful:

The logarithmic nature of the data matches how our ears perceive sound - a change of 10 dB is roughly perceived as “twice as loud.” So it also makes it easier to handle the huge range of sound intensities we encounter in the environment.

My sound level data is totally normal! The negative values just mean everything is related to the mic’s maximum capacity, which is good… or perhaps I need a much better mic?.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

Recently I have adquired a Raspberry pi 4 and decided to upgrade my BirdNet-Pi station Entrelomas to BirdNET-Go in the same location.

Real-Time Soundscape Analysis: Exploring BirdNET-Go and Custom Data Logging

I want to start with a massive shout-out to the developers who made this all possible. My deepest thanks go to Tomi P. Hakala, who built BirdNET-Go on top of the core BirdNET project, and to mcguirepr89 for his foundational work on the original BirdNET-Pi project, now directed by Nachtzuster on which I’ve relied on extensively.

Without the tireless efforts of these two incredible developers and all the many project contributors! my journey into birding and biodiversity monitoring as a citizen scientist simply wouldn’t have been possible.

Introducing BirdNET-Go: A High-Performance Acoustic Monitoring Tool

BirdNET-Go is a high-performance, continuous classification tool written in Go that acts as a real-time soundscape analyzer for avian vocalizations. Leveraging Artificial Intelligence, it’s designed for efficiency and built on the powerful academic work of the BirdNET project, while drawing architectural inspiration from BirdNET-Pi, which I used extensively as in this post.

Sleek Interface and Critical Integrations

BirdNET-Go provides a clean, modern web interface where users can easily view detected birds, analyze their frequencies, and play back captured recordings.

It also includes a streamlined, resource-light analytics dashboard, compared with the one in BirdNET-Pi that takes much longer to load. I also love the posibility to export the data as a csv file.

However, the most interesting feature, especially for home users, is its seamless integration with Home Assistant via MQTT. It supports audio streams directly via RTSP, which is the standard protocol for streaming video and audio from common IP cameras. This allows users to turn existing security hardware into a powerful, real-time acoustic monitoring station.

The Game-Changer: Soundscape Context

Beyond simple species identification, BirdNET-Go includes a powerful feature called Sound Level Monitoring. This is crucial for understanding the acoustic context of the environment, which helps in filtering unwanted noise and provides invaluable sounscape data for researchers.

The system registers sound levels in 1/3 octave bands, following the ISO 266 standard, providing a detailed frequency breakdown of the soundscape. This data can be useful to calculate the Noise Criterion (NC) - level. The data is recorded as frequently as every ten seconds, offering continuous, advanced audio monitoring that gives a far richer picture of the overall soundscape than just a list of bird names.

Notice I selected to get data each 30 seconds and not each 10 as default.

The Data Challenge: Optimizing for a Home Server

While this Sound Level data is incredibly valuable, accessing it required a solution tailored to my home lab server, which is an old Dell laptop running Home Assistant in a virtual machine of Proxmox. By the way here I got the inspiration to build it.

Sound level data from BirdNET-GO can be published via MQTT or accessed through a Prometheus-compatible endpoint, and it’s typically consumed by databases such as InfluxDB for high-resolution time series, and for its visualization is common to use tools like Grafana.

But dedicating that much processing power and space to a separate database in InfluxDB and for Grafana to visualize wasn’t a viable option for my Raspberry-Pi 4 or my old laptop as home server. To keep the resource load minimal, I decided on a low-tech, simple and diferent approach:

I leveraged BirdNET-Go’s tight integration with Home Assistant and created a simple automation to capture the necessary sound level attributes and write them directly to a local CSV file. This file now serves as my lightweight data repository, ready for deeper analysis and visualization using R.

The inspiration to integrate Home Assistant with BirdNET-GO came from Kyle Niewiada.

This is what you need:

1. Setup:

Save this Python script to: /config/scripts/save_audio_data.py.

This script reads the JSON data embeeded in the MQTT, extracting the time, frequency, noise, mean and maximum data, and save it as a new row in the soundlevel_data.csv file. Notice the file path at /config/www/.

2. Add this to configuration.yaml in Home Assistant

3. Add this automation in Home Assistant.

For that go to: Settings → Automations → Create Automation → Edit in YAML, then paste the automation code.

Notice the topic: birdnet-go/soundlevel should be the same topic you configure in BirdNET-GO. Also notice the last line includes inside the “” the word trigger.payload surrounded by double bracket {}. For some reason GitHub prevent to visualize it properly.

4. Restart Home Assistant

Voila!!! you have a csv file that grows by twenty rows each minute storing sound levels.

To download the csv data

I use WinSCP to make the file trasnsfer from and to my home server.

Of course BirdNET-GO is still a work on progress and its down side is that if you use an older, or diferent browser to Chrome, the visualization can freeze.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

The University of Maryland’s Global Land Analysis and Discovery (GLAD) laboratory, in collaboration with Global Forest Watch (GFW), has released its latest global forest loss data. This annually updated dataset, which you can download here, provides a unique look at how forests are disappearing around the world.

Using Landsat satellite imagery, the data reveals detailed spatiotemporal trends—showing not just where and when forest loss is happening, but also the rate at which it’s occurring. This helps researchers, policymakers, and the public understand the dynamics of global deforestation and its impact on our planet.

Although this is a global dataset, I wanted to apply it to a specific area of interest: Los Katíos National Park in Colombia. This is a site where our team is actively working, installing camera traps to monitor wildlife. We’ve observed a recent increase in deforestation here, making it a perfect case study.

My goal was to use this deforestation data as a covariate in an occupancy analysis. This approach allows us to investigate the relationship between forest loss and the presence or absence of wildlife species, providing a more robust understanding of how habitat changes impact local ecosystems.

This also provided the perfect opportunity to test the new animation features of the tmap package.

Below is R the code I used to download the boundaries of a national park using the wdpar package. This park boundary is then used as the area of interest (AOI) for the deforestation analysis with the gfcanalysis package.

Get the Data and Make the Analysis

This part download the park limit and makes the deforestation analysis.

library (gfcanalysis) # get Hansen data
library (sf) # make sf maps in R
library (wdpar) # get data from World Database on Protected Areas (WDPA) 
library (tidyverse) 

# directory to download Hansen Files
data_folder <- 'C:/tmp/Hansen_downloads/Katios'

# get all protected areas in Colombia
col_raw_pa_data <- wdpa_fetch(
  "COL", wait = TRUE, download_dir = data_folder
)

# clean Colombia data
# be patient... takes some time...
col_pa_data <- wdpa_clean(col_raw_pa_data)
# filter Katios National Park
Katios_NP <- col_pa_data %>% filter(NAME == "Los Katíos")

# add a buffer around to get a better context
buffered_Katios_NP <- st_buffer(Katios_NP, dist = 15000) # Buffer by 15 km 

# Area of interest --- Katios National Park sf file
aoi <- buffered_Katios_NP

# Calculate the google server URLs for the tiles needed to cover the AOI
tiles <- calc_gfc_tiles(aoi)

# Check to see if these tiles are already present locally, and download them if 
# they are not.
download_tiles(tiles, data_folder)

# Extract the GFC data for this AOI from the downloaded GFC tiles, mosaicing 
# multiple tiles as necessary (if needed to cover the AOI), and saving  the 
# output data to a GeoTIFF (can also save in ENVI format, Erdas format, etc.).
gfc_data <- extract_gfc(aoi, data_folder, 
                        filename="C:/tmp/Hansen_downloads/Katios/Katios_extract.tif",
                        overwrite=TRUE)

###############################################################################
# Performing thresholding and calculate basic statistics
###############################################################################

# Calculate and save a thresholded version of the GFC product
gfc_thresholded <- threshold_gfc(gfc_data, 
                                 forest_threshold=90, 
                                 filename="C:/tmp/Hansen_downloads/Katios/Katios_extract_thresholded.tif",
                                 overwrite=TRUE)

# annualized layer stack of forest change
Katios_annual_stack <- annual_stack(gfc_thresholded)

Katios_annual_stack is a raster brick from the old raster package. So in the next part we convert that object to a SpatRaster object using the terra package and use that SpatRaster “stack” as input for the animation.

Make the Animation

library(tmap)
library(tmaptools)

# convert rasterbrick to terra
Katios_annual_stack_t <-  terra::rast(Katios_annual_stack)

# make the map and animate
Katios_deforest <- tm_shape(Katios_annual_stack_t) + 
  tm_raster(palette  = c("#008000", "#ffa500", "#ff0000", 
                    "#0000ff", "#ff00ff", "#c0c0c0", "#101010"),
            style = "cat",
            labels = c("Forest", "Non-forest", 
                       "Forest loss", "Forest gain", "Loss and gain", "Water", 
                       "No data")) +
  tm_animate(fps = 1) +
  tm_shape(Katios_NP) +
  tm_lines(
    #lwd = "strokelwd", 
    lwd.scale = tm_scale_asis(values.scale = .5),
    col = "black") 

# save animation
tmap_animation(Katios_deforest, "C:/tmp/Hansen_downloads/Katios/Katios_deforest_1fps.mp4")

We can save the animation as .gif as well.

Here the Animation as a Video

It’s disheartening to see the rapid rise in deforestation since 2015, especially the dramatic explosion of forest loss within Los Katios National Park in 2021. This trend is a clear and alarming threat to the park’s biodiversity and ecological integrity.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

Al final de la pandemia en 2023, y como parte de mi entusiasmo del proceso de aprendizaje en métodos de monitoreo acústico que surgió con el proyecto Monitoreo de Biodiversidad en destinos de turismo de naturaleza de Awake Travel, me embarque en el proyecto de construir mi propio sistema de monitoreo acústico de aves con un BirdNET-Pi, siguiendo el tutorial de Core Electronics.

El aparato

Aprovechando una de las promociones de sigma electronica, adquiri un Raspberry Pi 3 Modelo A+ a muy buen precio, junto con un microfono USB.

Siguiendo el tutorial, el montaje y la instalación del BirdNET-Pi fue muy sencilla!

En un comienzo lo instale en la ventana de mi apartamento, pero muy pronto me di cuenta que solo estaba registrando a la gente, los perros y los carros, ya que la ventana daba a la parte interna y el parqueadero de los edificios. Así que decidí moverlo a la ventana del apartamento de mis suegros, que da hacia la parte exterior de los mismos edificios en Cajica, donde hay algunos potreros con vacas y cercas de pinos. Con esta nueva localización el dispositivo quedo registrado como la estación 754 en birdweather que recientemente movi a una nuevo sitio cercano registrado como la estación 18147 en birdweather.

Junto con la instalación inicial del BirdNET-Pi me di a la tarea de crear una cuenta en twitter llamada BirdNET Pi Cajica, para registrar las detecciones de forma automática, pero desafortunadamente, con la compra de twitter y su conversión a X, las políticas de las API para crear cuentas automatizadas cambiaron y la cuenta quedo desactivada luego de unos pocos meses de creada.

Teniendo en cuenta que los datos habían sido almacenados por birdweather me di a la tarea de recuperarlos usando las API de birdweather. Con esto he aprendido un poco sobre APIs y he logrado recuper los registros más comunes con un código muy sencillo. Acá se muestran los registros de las 15 especies más comunes con su nombre en español.

### R API to birdweather
library(httr)
library(jsonlite)
library(tidyverse)
library(kableExtra)

# 1st request. Ask who is
res = GET("https://app.birdweather.com/api/v1/stations/{API-KEYword-del-dispositivo}/species?period=all&locale=es")
res # to see 200 is success 

# from Jsaon convert to list
data = fromJSON(rawToChar(res$content))
names(data)

View(data$species)
names (data$species)

# make table with thumb
tbl_img <- data.frame(
  Common_name =c(data$species$commonName[1:15]),
  Scientific_name = c(data$species$scientificName[1:15]),
  Image = ""
)

# make a table
tbl_img  %>% kbl(booktabs = T) %>%
  kable_paper(full_width = F) %>%
  column_spec(1, color = "white",bold=T,
              background = c(data$species$color[1:15])) %>% 
  column_spec(2, #color = "white",
              italic =T,
              link=data$species$imageUrl[1:15]) %>% 
  column_spec(3, image = spec_image(
    c(data$species$thumbnailUrl[1:15]), 100, 100)) 

Extrayendo mas detalles de una especie

Para obtener más datos de una especie se puede usar el Get detections que acoplado a los paquetes de R httr y jsonlite, hacen que el manejo de la API dsde R sea muy fácil.

Para este ejemplo extraeremos las detecciones del mes de Marzo del 2023 y usando la magia del paquete camtrapR vamos a obtener una gráfica del patrón de actividad de ese mes.

library(httr)
library(jsonlite)
library(tidyverse)
library(camtrapR)

#  get turdus detections and select a few columns
turdus_mar <- GET("https://app.birdweather.com/api/v1/stations/{API-KEYword-del-dispositivo}/detections?from=2023-03-01&to=2023-03-31&speciesId=422&locale=es")
turdus_mar_data = fromJSON(rawToChar(turdus_mar$content))
turdus_mar_detections <- as.data.frame(turdus_mar_data$detections)|> 
  dplyr::select(c(id, timestamp, confidence, probability, score, certainty)) |> 
   mutate(Species="Turdus fuscater")

# fix datetime stamp
turdus_mar_detections$DateTime <- str_replace(turdus_mar_detections$timestamp, "T", " ") |> 
  str_sub(start=1,end=19) |> strptime ("%Y-%m-%d %H:%M:%S")

########### Activity Plot
activityDensity (recordTable = turdus_mar_detections,
                 species     = "Turdus fuscater",
                 recordDateTimeCol="DateTime") 
rect(0, 0, 6, 0.9, col= rgb(0.211,0.211,0.211, alpha=0.2), border = "transparent")
rect(18, 0, 24, 0.9, col= rgb(0.211,0.211,0.211, alpha=0.2), border = "transparent")

Es una lástima que las detecciones tengan un límite de 100 registros. Tal vez para hacerlo más detallado habría que bajar los datos diariamente.

El entusiasmo con el monitoreo acústico no ha parado. Recientemente instale un dispositivo igualito en la casa de mis padres en Cúcuta. Pero esta vez usando el nuevo código de BirdNet implementado en GO. Si bien tiene menos “features” y no viene con apraise, me parece que funciona bastante bien, siendo ligero y eficiente.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

Using the power of Quarto and R Quarto and R, I’ve embarked on a new project: a blog for the Diego of the future (and anyone else!) exploring the exciting possibilities of camera trap data analysis.

Content

This blog will help extract insights from camera trap data and provide additional resources for researchers and enthusiasts alike.

• I will give you some tips to improve data organization and accessibility, as well as guides for analyzing camera trap data.
• I will discuss techniques for removing errors, inconsistencies, or duplicates, emphasize the importance of metadata for understanding data context and provenance.
• Also discuss techniques for predicting species distribution based on environmental factors, discussions and tutorials on how to analyze diurnal or seasonal activity patterns.
• Introduce various R packages, and statistical models for analyzing camera trap data.
• I will try to discuss common challenges encountered in camera trap data analysis and potential solutions.

Why Quarto and R?

Quarto and R are amazing tools for content creation, and both fit very well in my most recent workflow. Thi new blog is organized with thematic posts giving a small introduction and providing a step-by-step code in R.

Language

The new blog is in English, however using the magic of google chrome you can translate it to Spanish or to any other language.

Next steps

This new blog is the first step in a planned evolution for my web content creation. I’m excited to explore the new possibilities that Quarto and R offer for web creation. I hope you enjoy the new content and please stay tuned for more exciting developments in the near future.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

GitHub has become my central platform for everything code-related. From development and sharing to version control and archiving, it’s my go-to tool. It even extends to web development and report management. Recently, I hit a milestone – over 100 repositories! While many are forks or clones of others’ work, it showcases the collaborative power of the open-source community that GitHub fosters. I’m incredibly grateful for this platform after a decade of using it!

Weppage update

After a whirlwind ten years, I’ve finally given my webpage a much-needed makeover! While I initially considered making a significant leap to Quarto and R for content creation, I ultimately decided to stay true to the platform that’s served me so well: the user-friendly (more less) Minimal Mistakes framework. Its familiarity and ease of use won me over once again.

The big change

The update involved a major shift in the comment section. I’ve transitioned from Disqus to Facebook for comments. This allows for a more streamlined and potentially more engaging user experience. However, there’s a trade-off: any comments previously made on the pages themselves are no longer visible. Thankfully, they’re not entirely lost – you can still access them on the Disqus platform if you wish. I am going to miss the coments to the Mammals of Colombia post and The ULEAM post. This decision wasn’t easy, but the recent changes to Disqus’s icons and reactions just weren’t aligning with the overall aesthetic I envisioned for the website. Perhaps in the future I can return to Disqus. Let see what will happen using facebook.

Next steps

This update is just the first step in a planned evolution for my web content creation. I’m excited to explore the new possibilities that Quarto and R offer for web creation in the near future. Perhaps in the next iteration, I can marry the user-friendliness of Minimal Mistakes with the power of these data science tools. In the meantime, I hope you enjoy the fresh look and feel of the site! Stay tuned for more exciting developments in the near future.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

Cursos

El curso de modelos de ocupación, en su version corta, se movio a la universidad de los Andes en Bogota gracias a la amable invitacion de la profesora Aida Otalora-Ardila https://dlizcano.github.io/Uniandes/.

Monitoreo de biodiversidad y turismo de naturaleza

En el 2023 se cerró el proyecto de monitoreo de biodiversidad con Awake en el cual tuve la oportunidad de viajar a lugares increíbles en Putumayo, Cauca, Choco y Casanare para hacer monitoreo acústico de biodiversidad y fototrampeo.

Posteriormente tuve la uportunidad de vincularme al programa de Destino Naturaleza de USAID en el cual pude aprender más de la realidad del turismo de naturaleza en Colombia y su fuerte conexion con la conservación de la biodiversidad. Conoci personas maravillosas, hubo muy buenos viajes y fue una estupenda experiencia.

Por otra parte

Gracias a la invitacion de Audubon Panama tuve la oportunidad de explorar con mucha profundidad mas aspectos del monitoreo acustico usando inteligencia artificial, esta vez en los manglares de la bahía de Parita en Panama. Pronto codigo, analisis y resultados…

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

Trampas Camara

Como parte de mi trabajo en Awake Travel realizamos una serie de videos sobre trampas camara que hacen parte del curso de fototrampeo la plataforma de AwakeU.

Fototrampeo para turismo de naturaleza

Cuantas camaras debo usar?

Como instalar las Camaras?

Cuanto tiempo debo dejar las camaras activas?

Que gran diferencia entre el primer video de la serie de acústica y el ultimo de la de fototrampeo!

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.