The sun is an energy resource that distributes its bounty unevenly across the Earth's surface: Some places are just sunnier than others. While the sunniness of a given region is pretty well established as a matter of popular lore, such assessments generally are not regarded as engineering-quality data.
Thomas Stoffel, manager of the National Renewable Energy Laboratory's (NREL) Resource Information and Forecasting Group in Golden, Colo., says successful solar power generation programs require scientific solar resource assessments. In particular, utility-scale projects that form the foundation of many national renewable energy policies around the world demand intense data collection and analysis over a period of years, if not decades.
‘There are 20 places in the world with the 10 years' worth of data I would trust,’ Stoffel says.
Which leaves a lot of ground to cover.
Solar resource assessment is a much different activity than solar forecasting, he says. The latter is essentially weather forecasting with an eye on evaluating the amount of sunlight expected over the coming hours and perhaps few days for use in planning how existing solar power generation facilities in a region might best be incorporated into the electricity grid to meet demand.
Solar resource assessment, on the other hand, is more akin to oil or natural gas exploration in that the sun's radiation reaching the ground in a region is measured as a means to determine optimal siting and configuration of solar power facilities that could operate for decades.
NREL's Solar Radiation Research Lab has been collecting precise measurements of the sun's local intensity since 1981, as well as examining methods for studying and presenting the data for use in resource assessment activities. Data collection involves an array of radiometer stations situated over the lab's research area on South Table Mountain.
Because it is not yet practical to cover the Earth in sensors, NREL produces a number of software tools that enables users to make a general assessment of the potential solar resources of a particular region. The tools are mostly presentation systems for viewing data collected over time, often many years. Essentially, you have a heat map of surface sun radiation that can be interrogated for specific figures for a given location.
‘All the maps are based on satellite data,’ Stoffel says.
The information comes from weather satellites, to be precise – the same ones that supply the National Weather Service with data and images for TV news. NREL has developed algorithms for taking the brightness from the images and converting it into radiation values, ‘with uncertainties,’ he notes.
By using weather reports, it is relatively easy for even a non-specialist to identify those portions of the Earth where sparse annual cloud cover should yield productive sites for solar power plants. However, Stoffel says that there can be tremendous variation in solar radiation within a region due to vagaries in local conditions that do not necessarily show up on weather maps. The variation is even measurable across the acreage of a given proposed site.
Thus, while weather maps with applied solar radiance algorithms may provide a high-level overview of the suitability of a proposed site, the attending uncertainties may be unacceptable to people with a stake in the actual project. Engineers, for example, require more extensive data points in order to site and design the installation.
As for those who actually have to finance a solar project, ‘bankability’ is how Stoffel measures the level of required quality of solar radiation data.
‘If I'm actually going to finance a project, I want more certainty than I can achieve from weather maps,’ he says.
NREL serves a number of important functions with its solar resource assessment support activities at home and abroad. Personnel identify required instrumentation – chiefly radiometers and attending data-networking equipment such as satellite communications or cellular modems. The organization assists in the design of research installations and configuration of instruments, and provides training and other support functions. Ultimately, the goal is to get the operators up and running with solar assessment facilities for themselves.
The Kingdom of Saudi Arabia has ambitious plans to install more than 50 GW of renewable energy in the next 20 years, with solar power meeting at least 30% of its electricity needs by 2032. The kingdom has turned to NREL for help in assessing its solar resources.
As part of the program, NREL is setting up between 75 and 100 measurement sites, with each site having dozens of measurement stations spread out over the study area. The stations, which collect data in one-second intervals, are intended to operate for several years. The longer and more detailed the observations, the more bankable the data. Stoffel says a year or three is needed to give engineers confidence.
‘We're doing for Saudi Arabia what I wish we were doing for the U.S.,’ he says.
However, there are statistical shortcuts for those without Saudi Arabia's deep pockets. NREL has turned its decades of observations in Colorado into a baseline measurement system. Data from the system is correlated with recorded weather data to produce models. Ideally, a prospective solar plant operator could use the models in concert with less stringent physical data collection to produce useful solar resource assessment reports.
The bankability of such reports, however, will be up to the financiers.