Impacts of long-timescale variability in solar resources at high PV penetrations: Quantification

Quantification and mitigation of solar resource variability´s impacts at short timescales (minutes to hours) have been examined extensively. This is because resource and subsequent solar generation variability as a result of short-term weather and diurnal effects poses the clearest link to utility planning timeframes: regulation markets, load following and unit commitment. Intermittency on the scale of days to weeks represents a longer-term, but no less important barrier to high-level solar penetration because it determines the types of solutions needed to ensure resource availability-primarily energy storage and long-distance interconnection. Using 24 years of globally distributed, daily-averaged, satellite-derived surface and top-of-atmosphere shortwave downward radiative flux data from the International Satellite Cloud Climatology Project (ISCCP) via NASA´s Surface meteorology and Solar Energy (SSE) database, we examine the relationships between surface global horizontal solar radiation variability and time-averaging interval length for time frames longer than one day. From these relationships, we discuss the methodology necessary to estimate the marginal increase in the cost of solar electricity needed to meet a synthesized flat load curve. Variability of is quantified at a given timescale by the standard deviation of all step changes in insolation between any two consecutive time-averaging intervals. We also investigate the impact of geographic footprint, from regional to continental scale, on this long-term variability metric and discuss how this can be used to calculate the marginal increase in the levelized cost of solar electricity. By analyzing these metrics for every time-averaging interval length from 1 day to several months, we observe a sinusoidally modulated exponential decay in the variability metric as a function of time averaging interval length.