The International Geosphere-Biosphere Programme (IGBP) is taking the first steps towards attaining a better understanding of the effects of such biome perturbations. The IGBP's data and information system is part of a program to make global data available to the scientific community. The main goal of the Global Primary Production Data Initiative (GPPDI) is to establish a baseline of net primary productivity (NPP) data from existing field data sets from grasslands and other biomes such as tropical rainforests and northern coniferous forests. ("Net primary productivity" refers to the total amount of new plant material produced by plants during photosynthesis.)

"Grassland systems are among the easiest in which to measure net primary productivity," said Richard Olson, Staff Scientist at Oak Ridge National Laboratory. Scientists measure net primary productivity because they want to know how much carbon and energy is transferred between the atmosphere and green plants, which is an indicator of the condition of the land. Net primary productivity is also a likely indicator of the response of ecosystems to rising atmospheric carbon dioxide and global climate change.

Olson used a study of the Central Plains Experimental Range (a Long-Term Ecological Research site) by William Parton at Colorado State University to compare the accuracy of different net primary productivity models. Thirteen different grassland sites were available from which comparison of the accuracy of different net primary productivity models could be estimated. "These 13 data sets gave us a good starting point. From there, we spent some time standardizing the documentation between the different data sets, but it seemed an easy place to begin data collection for the archive," said Olson.

Olson is heading the Oak Ridge component of the Global Primary Production Data Initiative effort to find, validate, and archive existing net primary productivity data for various global biomes by pulling together existing measurements, literature, and unpublished reports available worldwide. "Our job is to organize the data so that they are fairly consistent and representative of biomes worldwide," said Olson.

Data collection begins with a literature search by Olson and Oak Ridge Postdoctoral Fellow Jonathan Scurlock. Any small bit of pertinent information gleaned from published papers leads to correspondence with the principal investigator. "By establishing a good working relationship with the principal investigators, we are increasing our data catalog," Olson said. "Not only do we frequently receive the initial data set but we also receive data updates for the archive."

Next, Olson and Scurlock sort through and visually check the incoming data. Questions arising from checking the data and metadata are answered either in the accompanying literature or by the principal investigator.

Long range plans include scaling up these data sets to accurately represent 50 x 50 kilometer sites. This work began in late 1996 at the University of Maryland. Scientists use climate and vegetation data from the Global Primary Production Data Initiative study sites as inputs into their computer models that simulate the cycles of exchange of gases and nutrients between plants and their environment. Such models enable scientists to better estimate net primary productivity on regional scales. Ultimately, complete, regional scale data sets will be available for incorporation into scientists' models and research efforts to produce better models of global processes.