Posts in R

Snow Melt Forecasts for Mt. Rainier National Park

I'm spending some time at Mt. Rainier National Park this spring as a "Scientist in Residence", a position made possible by support from the Northwest Climate Science Center. One of the projects that I'm working on is generating up-to-date forecasts of when Mt. Rainier's (massive) snowpack will finally melt this summer. The timing of snow melt is important for wildflower phenology (one of my research interests), but it's also important because it determines when some of the park's management activities can happen (i.e. trail maintenance, road openings), and potentially for visitors. This post serves as a brief FAQ to the forecasts, which are available live today.

The link to the most up-to-date forecast map is here.

written in GIS, MapBox, R, maps Read on →

Easy Web Maps With R and Leaflet

I've been using R as a Geographic Information System (GIS) for a long time, and I've relied heavily on the excellent rgdal, maptools, and raster packages to process spatial data. One thing has always bugged me about this workflow, however. If I want to check or explore geographic data by clicking around on a map, I nearly always have to export it and open it up in another program like ArcGIS or QGIS. That's why I was really excited to learn that the folks at Rstudio have come up with a cool solution: they've written R wrappers for the leaflet javascript library which allows folks to quickly overlay vector data on an interactive web map.

In this post I'll describe how to build this map and share it on Rpubs or a self-hosted website. This assumes you have some familiarity with R and manipulating spatial data. For a great compendium of resources on GIS in R, see here

written in GIS, R, leaflet, maps Read on →

Phenology and the Intersection of Two Curves

My lab has been working on a project forecasting how the timing of flowering will vary over space and through time with climate change. To do this well, we need robust ways to quantify shifts in phenological timing. The approach we have taken so-far is to fit relatively simple functions to the observed data, but how do we quantify the amount of shift? Here's where numerical integration comes in. Here's an example with some simulated data. I've borrowed some of this code from here.

written in Phenology, R Read on →