The dataset contains plant species occurrence records in the upper to mid Tana River Basin, digitized from preserved specimens at the East African Herbarium (EA), Nairobi. The dataset is a subset of the EA BRAHMS database generated over several years through various projects and day-to-day digitization efforts. Specimen records from the Tana River Basin were extracted from the EA BRAHMS dataset then cleaned and formatted in Darwin Core Standard. Data cleaning was done using Open Refine Software as well as the GPS Visualizer tool to generate altitude values from coordinates. Elevation data from GPS Visualizer is controlled from a variety of reputable sources including SRTM, ASTER, USGS and Google Maps. Other cleaning tasks included correcting spellings, filling in missing data, checking the taxonomy, dates, and subnational administrative areas. Records without coordinates were georeferenced using GEOLocate, Google Maps, and Gazetteers. The field altitude (m) was automatically generated from GPS Visualizer (https://www.gpsvisualizer.com/elevation) to generate missing elevation values. The process involved uploading a CSV format file with latitude and longitude values to the website which then sampled these coordinates against inbuilt elevation profile files (Digital Elevation Models) to give altitude values at specified coordinates. Most coordinates were also retrieved from gazetteers as they are legacy data.