Peru is one of the most biologically diverse countries in the American tropics. Ecosystems range from coastal mangroves and deserts to high-elevation Andean grasslands and cloud forests, submontane Andean forests, and the vast jungles and palm swamp wetlands of the Amazonian lowlands. The country harbors more than 40,000 species of plants, of which more than 5000 are endemic. Despite decades of botanical exploration and plant collection in many regions of the country and long-term development of numerous herbaria, small and large, no major effort has been made toward the integrated digitization of national plant specimen holdings. Therefore Peruvian herbaria remain disconnected from foreign herbaria that include Peruvian plant collections in existing online databases. Moreover, the rich diversity in plant specimen holdings of those herbaria is still inaccessible to the scientific community and general public. This is a shame since many Peruvian herbaria have collections that are historically important and not represented in the many existing online herbaria from around the world.
Key Objectives and Activities
The goal of this project is to digitize up to 80,000 Peruvian plant specimens and make the data and images available through an innovative online community-driven information system that already hosts thousands of plant specimens from Peru. While the digitization of all Peruvian herbaria is beyond the scope of this phase, the project team includes colleagues from three major Peruvian herbaria, all of whom have agreed to make immediate advances toward the digitization and dissemination of their collections. These are the: (1) Forestry Herbarium of the Universidad Nacional Agraria La Molina – MOL, (2) the Vargas Herbarium of the Universidad Nacional San Antonio Abad del Cusco – CUZ, and (3) the Herrense Herbarium of the Instituto Nacional de Investigaciones de la Amazonia Peruana – IIAP. The planned activities include:
Acquire and install herbarium digitization equipment.
Train students and staff in the herbarium digitization work flow.
Implement a workflow that integrates the selection, curation, and imaging of herbarium specimens as well as the digitization of label data.
Image 80,000 specimens and import data and images into Atrium.
Prepare example field guides through an automated prototype publication system powered by Atrium.
Planned Outputs and Outcomes
The project will produce 50,000 new high resolution specimen images and make the images and data available to the general public through an online data portal. The resulting data will be an important contribution to the knowledge of the flora of the Andes-Amazon region and other areas of Peru. Because many of the plant specimens to be imaged come from high priority biodiversity hotspots in Peru, the project will have broad impacts on conservation-driven research and synthesis in the region. Such outcomes will include analysis and synthesis at taxonomic, ecosystem, local, and regional levels. The planned work flow will also contribute to the training of an impressive number of Peruvian students, herbarium curators, and staff in the herbarium digitization protocol. At the same time, the process of selecting and organizing specimens that pass through the imaging system will include a step of cleaning, repairing, re-gluing, or re-labeling some specimens; all important work that will result true renovation of three important herbaria of Peru. In the end, the act of making thousands of high-resolution plant specimen images available through an online information system is hoped to decrease the distance between discovery and documentation of the rich flora and ecosystems of Peru.
Results to Date
To date the team has designed and installed imaging equipment based on a custom light box, a sturdy stand, and a Nikon D800E DSLR camera in three Peruvian herbaria. An integrated work flow has been implemented to incorporate specimen curation and imaging, image processing, label data capture, and data upload and sharing. Herbarium staff and students have been trained in the complete process of herbarium digitization.
The Herense Herbario of the Jenaro Herrera Biological Station outside of Iquitos in the northern Peruvian Amazon has been completely digitized. More than 11,000 specimen images representing more than 6000 collections have been imaged and shared through the online biodiversity information system.
Approximately 20,000 specimens from the Forestry Herbarium of the Universidad Nacional Agraria La Molina of Lima have been imaged. To date approximately 12,000 have been made available through the online biodiversity information system.
Over 20,000 specimens from the Vargas Herbarium of the Universidad Nacional San Antonio Abad of Cusco, Peru, have been imaged to date. Approximately 12,000 of those specimen images are available through the online biodiversity information system.
Since the beginning of 2014 more than 3000 new users have visited the online biodiversity information system and utilized the data and images produced through this project. About 25% of those new users are Peruvians, which attests to the demand that exists for accessing botanical information from within the country. The success of this project has led to an increased interest by other Peruvian herbaria to digitize their collections. Most importantly, these collections that were only accessible to experts on site are now digitized and available to Peruvian and global researchers. This work has also contributed analyses and publications that have stimulated a new interest from policy makers in major wetlands conservation.
De Stefano, RD; JP Janovec; LL Can (2013) Three decades to connect the sexes: Calatola microcarpa (Icacinaceae), a new species from the Southwestern Amazon. Phytotaxa. 124 (1): 43–49. (link)
Janovec, JP; JE Householder; M Tobler; et al. (2013) Evaluación de los Actuales Impactos y Amenazas Inminentes en Aguajales y Cochas de Madre De Dios, Perú. (Evaluation of the Current Impacts and Imminent Threats on Aguajal and Oxbow Lake Wetlands of Madre de Dios, Peru.) 244 pages, 180 color image and map figures. World Wildlife Fund (WWF), Lima, Peru. (LINK)
The project team has learned important lessons in the implementation of a productive herbarium digitization work flow in Peruvian herbaria. The design, installation, and calibration of the innovative imaging system required more time than predicted. The work flow requires not only specimen imaging and databasing, but also the curation and refurbishing of many specimens before they can be imaged. In addition, extra training and more focus than planned has been required to capture metadata from plant specimen labels. The challenges faced led to reducing the goal from 80,000 images in one year to 50,000 in 1.5 years. However, those numbers don’t tell the story of the emergent interest in digitization on major Peruvian herbaria and a new interest in data sharing and access.
Primary Software Platforms
All information from this project is integrated with tens of thousands of existing plant specimen records and images in the Atrium Biodiversity Information System for Peru http://atrium.andesamazon.org/
Atrium provides free global access to all the specimen data and images.
Atrium uses the IIPImage server to serve static and high-resolution zoomable images from a single JPEG2000 master file.
Collection data are served to the the Global Biodiversity Information Facility (GBIF).
Notes from JRS
This project, which ended in May 2014, has been an unqualified success in bringing digitization and data sharing to some of the most important plant collections in Peru. In addition to the sheer numbers of imaged and curated records, there is now human and technical capacity to continue this work. We hope this begins to create the basis of a dynamic digital flora of Peru including important ecosystems of the Peruvian Amazon, coastal mangroves, and dry forests, etc. The lessons learned from this project and others are contributing to new project designs in JRS grants that emphasize up-front analysis of collections quality and size and the testing of workflows early in project implementation.