My TEDx talk is now available on YouTube. The basic premise of the talk is that when you put all the recent advances in perspective, it results in a democratization of data on one hand, and science on the other. It means that anyone with a computer and a decent internet connection can have access to satellite data, process them using open source methods and extract information for their own use. This not only helps accelerate the rate at which scientific discoveries are made but also makes knowledge sharing and international development more equitable and inclusive.
Profile in a Swedish popular science magazine
Back in January, I had the opportunity to be interviewed by the Swedish popular science magazine Populär Astronomi about my research in Earth observation.
Perform a Mann-Kendall trend test on satellite image time series in R
The Mann-Kendall trend test has become popular in the remote sensing community to test whether a time series of satellite observations is consistently increasing or decreasing. In this post, I developed a function in R that can take in raster stacks or bricks to perform the Mann-Kendall trend test and calculate its statistical significance (p values).
Keynote Talk: “Earth from Space” at Astronomins dag och natt 2020
I was thrilled to be provided the opportunity to be a keynote speaker for Astronomins dag och natt or “Astronomy’s day and night” organized annually by the Swedish Astronomical Society. The event is in its 9th year and the theme for 2020 is Earth 2.0 and the intersection of the possibility of life in space and the possibilities that Earth observation offers to better understand our own planet.
Tutorial: Machine learning classification of Sentinel-2 satellite imagery using R [Updated]
Note: This tutorial was updated on April 20th, 2020 based on reader feedback.
In this short post, I would like to help you conduct your own machine learning classification of Sentinel-2 data using the open source package R. The process is pretty straightforward if you have experience in remote sensing and image classification. Even if you don’t have extensive experience, basic knowledge of remote sensing terminology is sufficient.
Test pixelwise correlation between two time series of gridded satellite data in R
Satellite time series data are useful for studying biophysical how variables change over time and understanding what causes those changes. Recently, I was looking into correlating two time series datasets over Africa to look at the relationship between net primary production (NPP) and rainfall. After a futile attempt to find an “out-of-the-box” software package that does this, I created an R function to speed things up.
Behind the Paper: Comparing machine learning algorithms using Sentinel-2 data
The field of machine learning is moving fast, and it seems that new fancy algorithms coming out every week. Sometimes, it is confusing to figure out which algorithms are best suited for which purpose. This is particularly the case when it comes to land-use and land-cover classification using multidimensional satellite imagery because most of the new algorithms are tested with either binary or uni-dimensional data.
Behind the Paper: Combining science and art to understand environmental change
A couple of years ago, my colleague and friend Emma Li Johansson approached me with an interesting proposition. Her idea was that the combination of remotely-sensed land cover and and land use change maps and participatory paintings of people’s perceptions of land change might provide useful information in understanding the drivers of change. She didn’t have to say any more because I was sold. It was a solid idea with an innovative interdisciplinary workflow and high potential to provide valuable outcomes.
Oil and opportunities for development in Somalia
Behind the Paper: Testing a new satellite-derived vegetation index in a new biome
The last chapter of my PhD dissertation was published earlier this year in the International Journal of Applied Earth Observation and Geoinformation. From conception to publication, this paper took about two-and-a-half years of work. It signifies the end of my PhD era, so to speak. The idea came from my supervisor, who suggested that testing the relatively new Plant Phenology Index (PPI) in semi-arid biomes would be a worthwhile cause because its only been evaluated in the boreal biome.