Printer Friendly

SOLAR RENEWABLE ENERGY: The Important and Challenging Role for Meteorology.

We see them everywhere, popping up like crocuses in spring: solar panels, newly installed on roofs, in fields, looking expectantly skyward for some sunshine. Solar energy production has more than quadrupled in the United States in the last five years alone, (1) as governmental incentives, personal preferences, and improved technologies continue to make it a more attractive choice of energy capture. What may not be as obvious are the number of hurdles that need to be cleared for solar renewable energy to be easily and readily absorbed into our electrical systems--no simple task. And the true surprise is that meteorology and climate science are at the center of some important issues to overcome.

To understand the complex role that meteorology plays with solar renewable energy, one must first understand the perspective of electricity suppliers. Our electricity companies need to balance supply and demand virtually instantaneously, with clear and strong penalties if they have too little electricity to meet demands. Often, they need to make commitments hours or even days in advance, which means if they don't know exactly how much energy they can count on from solar renewable energy, then they need to continue to have backup forms of energy--such as coal, nuclear, and natural gas--at the ready. The difficulty in forecasting clouds is well understood. Imagine having to forecast with a high level of accuracy at the one-minute level for the next day? The uncertainties of such a forecast can be very large, of course, meaning that our electricity providers will be using other sources of electricity production to assure they can meet demand. Improved forecasts of solar radiation along with reliable estimates of uncertainty on those forecasts are critical to being able to rely on solar energy.

As if the difficulty of forecasting clouds isn't daunting enough, one also needs to think more carefully about the collectors of solar energy. We see the flat, photovoltaic panels on roofs most everywhere. Understanding that forecasting solar energy for one of these installations requires being concerned with the angle of the panel and the direction it is facing, the forecasting task suddenly becomes more difficult. Now also consider whether the trees nearby--often taller than the panels--are full of leaves or have dropped their leaves for winter. Moreover, if it has recently snowed, it is important to know the timing of the melt off from the panels because that has a dramatic effect on power output levels. All of these issues of placement suddenly become very important to forecasting energy production.

Pushing beyond what we commonly see in our neighborhoods, major solar installations are setting up in less populated areas. These often rely on technologies quite different from the flat panels we are used to seeing. Using mirrors and other devices, sunlight can be focused into concentrated areas and then collected with higher efficiency. These efforts rely on the direct beam of solar radiation, which requires a different, more specialized prediction than forecasting for the flat solar panels we are used to seeing. Not only is the radiation at concentrated solar installations reaching the collectors at a more efficient angle than the radiation reaching flat panels, but also the solar collectors that convert the radiation can have direction collection efficiencies, with some taking in all radiation that can be absorbed as heat and others focusing on specific wavelength regions with highly specialized conversion materials.

Businesses have increasingly become involved in establishing solar renewable-energy collectors. In addition to direct marketing to homes and individual leases to residents, in the last few years businesses have sprung up to offer individual ownership of panels without having the panels installed on the individual's roofs. In these more concentrated cases, siting of these community solar farms becomes economically very important to the owners, with microclimates often determining a solar farm's level of success. As the price of the least expensive form of solar panels has plummeted in the last few years, more installations are being installed over larger and larger parts of the country, making them economically feasible in areas that, only a few years ago, were not considered viable.

Digging a little deeper into this issue, we need to start thinking in more detail about the sensitivity across the spectral range of the collector. The Department of Energy (DOE) and entrepreneurs worldwide are investing in technologies to increase the power production efficiency of the equipment while minimizing the cost. The diversity of technical approaches results in some sensors showing preferential sensitivity to different spectral ranges of the sun's radiation. For sensors sensitive to the longer wavelengths, water vapor can be more important than for sensors in the shorter wavelengths.

Additional factors become important when trying to understand the flow of energy from the sun to usable technologies. Photovoltaic cells work more efficiently at capturing photons in colder climates than in hot ones. This difference can be significant, because each degree in temperature that crystalline cells increase reduces electricity production by as much as a 0.5%. This effect can be large, because on a hot sunny day in summer in the midlatitudes, solar panel cells can exceed temperatures of 70[degrees]C, and so the electric production is reduced by 22.5% compared with the same panel at 25[degrees]C. Thus, forecasting solar renewable energy requires more than forecasting the amount of direct and diffuse solar radiation reaching the detector. Air temperature, wind speed, and even ground temperature can impact the efficiency of a detector by influencing its temperature.

Therefore, forecasting for solar renewable energy requires an understanding of the geometric, radiative, and temporal responses and operating parameters of the solar energy collectors. This may be daunting, particularly when one wants to forecast both magnitude and timing of energy output with high precision for a 24-hour period. Our understanding and ability to create accurate forecasts are limited at this stage. One factor working against progress is the lack of skillful cloud-resolving forecast models. A second factor is the lack of detailed, well-calibrated radiation measurements, particularly of the direct beam that is needed for various concentrated solar technologies.

Fortunately, a few factors are working in favor of improved forecasts. First, weather models continue to improve. Focused and collaborative efforts funded by the Department of Energy (DOE) and coordinated by NOAA's Office of Oceanic and Atmospheric Research have improved basic models that can help forecast solar radiation. Tailored models, often nested within NWS forecast models, are better than ever before, and plans are underway to make them even better with likely direct impacts on solar as well as wind forecasts. Second, with the large level of solar installations both in the United States and around the world, we have the opportunity to collect more data than ever before to develop, test, and refine our solar energy forecasts. And finally, for many applications we can balance errors. As Laura M. Hinkleman et al. showed in their 2011 presentation at the 2nd AMS Conference on Weather, Climate, and the New Energy Economy (https://ams .confex.com/ams/91Annual/webprogram/Paper186374 .html), if the amount of electrical production at one installation has been underestimated, this is often accompanied by its overestimation in a nearby installation. This balancing of errors can work best in a broken cloud situation, where the amount of sun on one homeowner's roof can be quite different from the neighbor's down the block. In this situation, forecasting for the neighborhood's production can be more accurate than for a single home, even if the installations' angles, direction, and tree coverage are quite different. Further, in a recent paper published in Nature Climate Change, Sandy MacDonald and Christopher Clack et al. showed the effects of geographic area on wind and solar deployment over the United States. For that paper, Clack developed a novel technique for estimating solar PV power by combining satellite, NWP assimilation, and high-precision ground-based irradiance data with statistical methods. The technique was necessary because of the imprecision of the NWP estimates. The paper demonstrated that with proper planning of wind and solar installations, along with transmission lines, carbon emissions from electrical generation can be reduced by 80% (compared with 1990 levels) without increasing costs.

Identification of improvements in solar cells often involves comparison of how those cells respond to a standard solar spectrum. Different locations have different amounts of ozone, water vapor, aerosols, and clouds, which can all affect the solar spectrum reaching the surface of the Earth (Fig. 1). Thus, the optimal solar cell may actually be location dependent. Economics may dictate both a range of solar collectors as well as a careful matching of solar collectors to climatological solar spectra. Currently, care is given to identify locations where concentrating solar plants will be viable. In the future, the matching of technology to solar radiation spectra may be more involved. Ongoing work funded by DOE and the detailed measurements at the Atmospheric Radiation Measurement locations may lead to new optimization for these more complex sitings. For concentrated solar installations, the integrity of the direct beam becomes the critical parameter to care about, bringing additional siting challenges.

The underlying observations supporting this research are currently being collected by a few different networks dedicated to monitoring solar radiation. Instruments that gather spectral information are more expensive and require more technical involvement. Instruments that measure solar radiation on a slanted surface, as would be needed to directly mimic the radiation received on slanted roof panels, are I quite rare. Instruments that measure the direct beam of solar radiation benefit from instrumental development of shadow-band radiometers. Intensive radiation sites, such as the Atmospheric Radiation Measurement program as well as the NOAA Integrated Surface Irradiance Study (ISIS) and Surface Radiation (SURFRAD) networks, combine spectral radiation measurements and direct beam observations with observations of the components of the atmosphere, including aerosols and water vapor. Figure 3 shows where this long-term solar radiation monitoring is taking place in the continental United States.

Addressing the meteorological concerns of solar energy will directly feed into its economic viability. While some may switch to solar energy no matter what the cost, many more individuals, companies, and electricity providers will only turn to solar energy if it's economically beneficial and meteorologically feasible. As individuals, and as a society, we'll be weighing the costs and benefits. Countries will be asked to hold to their commitments of last year's Conference of the Parties (COP-21), and the proper engagement of atmospheric scientists will be critical to economic alternatives to conventional energy. This year's AMS Annual Meeting is focusing on the role of observations; scarce and important solar radiation observations are among those being explored. As meteorologists, we have the unique role of addressing the outstanding issues regarding assessing, understanding, and forecasting solar energy.

ACKNOWLEDGEMENTS. Thanks to PV Lighthouse and Dr. Knut Stamnes of Stevens Institute of Technology for the information on spectral radiances, and to Orri Johnson at PV Measurements for the valuable information on spectral responsivity of solar cell materials. Photo of Skytrough parabolic trough collector courtesy of Skyfuel, Inc.

FOR FURTHER READING

Clack, C. T. M., 2017: Modeling Solar irradiance and solar PV power output to create a resource assessment using linear multiple multivariate regression. J. Appl. Meteor. Climatol, doi:10.1175/JAMC-D-16-0175.1, in press.

Dubey, S., J. N. Sarvaiya, and B. Seshadri, 2013: Temperature dependent photovoltaic (PV) efficiency and its effect on PV production in the world--A review. Energy Procedia, 33, 311-321.

Green, M. A., K. Emery, Y. Hishikawa, and W. Warta, 2011: Solar cell efficiency tables (version 37). Prog. Photovoltaics, 19, 84-92, doi:10.1002/pip,1088.

Hinkelman, L. M., R. George, S. Wilcox, and M. Sengupta, 2011: Spatial and temporal variability of incoming solar irradiance at a measurement site in Hawai'i. 2nd AMS Conf. on Weather, Climate, and the New Energy Economy, Seattle, WA.

King, D. L., J. A. Kratochvil, and W. E. Boyson, 1997: Temperature coefficients for PV modules and arrays: Measurement methods, difficulties, and results. 26th IEEE Photovoltaic Specialists Conf., Anaheim, CA.

MacDonald, A. E., C. T. M. Clack, A. Alexander, A. Dunbar, J. Wilczak, and Y. Xie, 2016: Future cost-competitive electricity systems and their impact on US CO, emissions. Nature Climate Change, 6, 526-531, doi:10.1038/nclimate2921.

Marquis, M., S. Albers, and B. Weatherhead, 2011: For better integration, improve the forecast. Solar Today, 25,52.

Perez, R., and T. E. Hoff, 2011: Solar resource variability: Myth and fact. Solar Today, 25.

U.S. Energy Information Administration, 2010: Renewable energy consumption and electricity preliminary statistics 2010. U.S. Department of Energy, 13 pp. [Available online at www.eia.gov/renewable/annual /preliminary/pdf/preliminary.pdf.]

Elizabeth C. Weatherhead (University of Colorado at Boulder) and Christopher T. M. Clack (Vibrant Clean Energy, LLC, Boulder, Colorado)

(1) 98 trillion Btu's in 2009 compared to 427 trillion Btu's in 2014 (U.S. Energy Information Administration; www.eia .gov/totalenergy/data/monthly/pdf/sec 10_3.pdf).

Caption: Fig. 1. Various existing technologies can make use of the different wavelengths in the solar spectrum. Understanding the spectral response of the solar collector is critical to estimating and forecasting solar renewable energy. Because of the solar spectra due to sun angle and atmospheric constituents, including clouds, water vapor, ozone, and particulates, different collectors can be more cost effective than others depending on location. Solar spectra are estimated for noontime radiation incident on solar cells angled for maximum radiation collection. The solar collectors in this figure make use of different chemicals for the fundamental transfer of photons to electrical energy: gallium indium phosphide (GalnP) is shaded in orange at the shortest wavelengths, gallium arsenide (GaAs) is shaded in yellow near 800 nm, and germanium (Ge) is shaded in pink for the longest wavelengths. Units are for standard radiation profiles and maximum percentage of absorption.

Caption: Fig. 2. Nineteen states now have some type of major solar installation. Often these dedicated installations use focused solar energy to convert sunlight into heat and then into electricity. Estimating and forecasting the direct beam of radiation used in these technologies is challenged by a lack of research and monitoring. Even for flat panels, optimizing the angle for installation depends not only on latitude but also climatology of atmospheric conditions, primarily cloud cover.

Caption: Fig. 3. Intensive solar radiation monitoring includes monitoring of the direct beam of radiation as well as some spectral resolution of the downwelling solar radiation. Three national networks coordinate the collection of these data: the Department of Energy's Atmospheric Radiation Monitoring site is shown in red, NOAA's SurfRad sites are shown in blue, and NOAA's ISIS sites are shown in green.
COPYRIGHT 2017 American Meteorological Society
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2017 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:NOWCAST: FORECASTING
Author:Weatherhead, Elizabeth C.; Clack, Christopher T.M.
Publication:Bulletin of the American Meteorological Society
Geographic Code:1USA
Date:Jan 1, 2017
Words:2424
Previous Article:A New Way to Forecast Extreme Events.
Next Article:MAKING CLIMATE DATA SING: Using Music-like Sonifications to Convey a Key Climate Record.
Topics:

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters