Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized JANUARY 2015 MODEL VALIDATION REPORT Solar Resource Mapping in the Maldives This report was prepared by GeoModel Solar, under contract to The World Bank. It is one of several outputs from the solar Resource Mapping and Geospatial Planning Maldives [Project ID: P146018]. This activity is funded and supported by the Energy Sector Management Assistance Program (ESMAP), a multi-donor trust fund administered by The World Bank, under a global initiative on Renewable Energy Resource Mapping. Further details on the initiative can be obtained from the ESMAP website. This document is an interim output from the above-mentioned project. Users are strongly advised to exercise caution when utilizing the information and data contained, as this has not been subject to full peer review. The final, validated, peer reviewed output from this project will be the Maldives Solar Atlas, which will be published once the project is completed. Copyright © 2015 International Bank for Reconstruction and Development / THE WORLD BANK Washington DC 20433 Telephone: +1-202-473-1000 Internet: www.worldbank.org This work is a product of the consultants listed, and not of World Bank staff. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work and accept no responsibility for any consequence of their use. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for non-commercial purposes as long as full attribution to this work is given. Any queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: +1-202-522-2625; e-mail: pubrights@worldbank.org. Furthermore, the ESMAP Program Manager would appreciate receiving a copy of the publication that uses this publication for its source sent in care of the address above, or to esmap@worldbank.org. World Bank, Global ESMAP Initiative, Renewable Energy Resource Mapping: Solar − Maldives (Project ID: P146018) Model Validation Report – Preliminary Results World Bank Group, Global ESMAP Initiative: Renewable Energy Resource Mapping: Solar − Maldives Project ID: P146018 Model Validation Report – Preliminary Results January 2015 GeoModel Solar, Pionierska 15, 831 02 Bratislava, Slovakia http://geomodelsolar.eu Reference No. (GeoModel Solar): 129-02/2015 Page 3 of 39 World Bank, Global ESMAP Initiative, Renewable Energy Resource Mapping: Solar − Maldives (Project ID: P146018) Model Validation Report – Preliminary Results TABLE OF CONTENTS Table of contents .....................................................................................................................................................4! Acronyms .................................................................................................................................................................5! Glossary ...................................................................................................................................................................7! 1! Summary ............................................................................................................................................................8! 2! Model quality indicators ...................................................................................................................................9! 3! Inventory of solar atmospheric and meteorological validation data .........................................................10! 3.1! Solar resource measurements .............................................................................................................10! 3.2! Solar resource modelled data...............................................................................................................11! 3.3! Atmospheric data..................................................................................................................................12! 3.4! Meteorological measurements .............................................................................................................13! 3.5! Meteorological models..........................................................................................................................14! 4! Validation of aerosol data ..............................................................................................................................16! 4.1! Evaluation of MACC-II Aerosol Optical Depth data ..............................................................................16! 4.2! Seasonal variability of Aerosol Optical Depth.......................................................................................18! 5! Validation of solar resource data ..................................................................................................................20! 5.1! Quality control of solar validation data..................................................................................................20! 5.2! Validation of solar resource model .......................................................................................................23! 6! Validation of meteorological data .................................................................................................................27! 6.1! Validation sites .....................................................................................................................................27! 6.2! Air temperature at 2 metres ..................................................................................................................27! 6.3! Relative humidity ..................................................................................................................................28! 6.4! Wind speed...........................................................................................................................................29! 6.5! Precipitation ..........................................................................................................................................31! 7! Uncertainty of the model estimates ..............................................................................................................33! 7.1! Solar resource parameters ...................................................................................................................33! 7.2! Meteorological data ..............................................................................................................................34! 8! List of figures ..................................................................................................................................................35! 9! List of tables ....................................................................................................................................................36! 10! References .....................................................................................................................................................37! 11! Background on GeoModel Solar .................................................................................................................38! Page 4 of 39 World Bank, Global ESMAP Initiative, Renewable Energy Resource Mapping: Solar − Maldives (Project ID: P146018) Model Validation Report – Preliminary Results ACRONYMS AERONET The AERONET (AErosol RObotic NETwork) is a ground-based remote sensing network dedicated to measure atmospheric aerosol properties. It provides a long-term database of aerosol optical, microphysical and radiative parameters. AOD 670 Aerosol Optical Depth at 670 nm. This is one of atmospheric parameters derived from MACC-II database and used in SolarGIS. It has important impact on accuracy of solar calculations in arid zones. BoM Bureau of Meteorology, national weather service in Australia BSRN Baseline Surface Radiation Network CFSR/CFSv2 Climate Forecast System Reanalysis (CFSR) and its operational extension, the Climate Forecast System Version 2 (CFSv2), are global meteorological models operated by the US service NOAA. CMSAF Satellite Application Facility on Climate Monitoring (CMSAF) aims at the provision of satellite-derived geophysical parameter data sets suitable for climate monitoring. Several cloud parameters, surface albedo, radiation fluxes at the top of the atmosphere and at the surface as well as atmospheric temperature and humidity products form a sound basis for climate monitoring of the atmosphere are available. DIF Diffuse Horizontal Irradiation, if integrated solar energy is assumed. Diffuse Horizontal Irradiance, if solar power values are discussed. DNI Direct Normal Irradiation, if integrated solar energy is assumed. Direct Normal Irradiance, if solar power values are discussed. ECMWF European Centre for Medium-Range Weather Forecasts is independent intergovernmental organisation supported by 34 states. The Centre provides operational medium- and extended-range global forecasts and a computing facility for scientific research. GAW Global Atmosphere Watch. It is a worldwide system established by the World Meteorological Organization to monitor trends in the Earth's atmosphere. GDPS (GRIB2) Global Deterministic Forecast System (GDPS, GRIB2) developed by the Meteorological Service of Canada (MSC). It contains data from analysis systems along with output from many of the Canadian Meteorological Centre's Numerical Weather Prediction (NWP) models. The data is free, under standard terms and conditions. GFS Global Forecast System. The meteorological model operated by the US service NOAA. GHI Global Horizontal Irradiation, if integrated solar energy is assumed. Global Horizontal Irradiance, if solar power values are discussed. GTI Global Tilted (in-plane) Irradiation, if integrated solar energy is assumed. Global Tilted Irradiance, if solar power values are discussed. MACC Monitoring Atmospheric Composition and Climate – meteorological model operated by the European service ECMWF (European Centre for Medium-Range Weather Forecasts) MMS Maldives Meteorological Service Page 5 of 39 World Bank, Global ESMAP Initiative, Renewable Energy Resource Mapping: Solar − Maldives (Project ID: P146018) Model Validation Report – Preliminary Results Meteonorm Meteonorm is a database with a monthly climate averages from meteorological stations around the world developed by Meteotest. This includes global radiation, temperature, humidity, precipitation, days with precipitation, wind speed and direction and sunshine duration. Data for any geographical location is calculate by spatial interpolation of values from the nearby meteorological stations. Meteosat IODC Meteosat satellite operated by EUMETSAT organization over Indian Ocean and Asia NCDC NOAA's National Climatic Data Center (NCDC) is responsible for preserving, monitoring, assessing, and providing public access to the climate and historical weather data and information. NCDC database contains mainly collection of data from the national networks belonging to World Meteorological Organization (WMO). NOAA NCEP National Oceanic and Atmospheric Administration, National Centre for Environmental Prediction. PVGIS Photovoltaic Geographical Information System developed by Joint Research Centre (JRC), the European Commission. Online free solar photovoltaic energy calculator for stand-alone or grid-connected PV systems. PVGIS works for Europe, Africa and Asia. Solar electricity generator simulation and solar radiations maps. RSR Rotating Shadowband Radiometer TEMP Air Temperature at 2 metres Page 6 of 39 World Bank, Global ESMAP Initiative, Renewable Energy Resource Mapping: Solar − Maldives (Project ID: P146018) Model Validation Report – Preliminary Results GLOSSARY Aerosols Small solid or liquid particles suspended in air, for example soil particles, sea salts, pollen or air pollution such as smog or smoke. Bias Represents systematic deviation (over- or underestimation) and it is determined by systematic or seasonal issues in cloud identification algorithms, coarse resolution and regional imperfections of atmospheric data (aerosols, water vapour), terrain, sun position, satellite viewing angle, microclimate effects, high mountains, etc. KSI Kolmogorov-Smirnoff index. It characterizes representativeness of distribution of high frequency (e.g. hourly) values. Root Mean Square Represents spread of deviations given by random discrepancies between measured and Deviation (RMSD) modelled data and is calculated according to this formula: On the modelling side, this could be low accuracy of cloud estimate (e.g. intermediate clouds), under/over estimation of atmospheric input data, terrain, microclimate and other effects, which are not captured by the model. Part of this discrepancy is natural - as satellite monitors large area (of approx. 3 x 3 km in Maldives), while sensor sees only micro area of approx. 1 sq. centimetre. On the measurement side, the discrepancy may be determined by accuracy/quality and errors of the instrument, pollution of the detector, misalignment, data loggers, insufficient quality control, etc. Solar irradiance Solar power (instantaneous energy) falling on a unit area per unit time [W/m2]. Solar resource or solar radiation is used when considering both irradiance and irradiation. Solar irradiation Amount of solar energy falling on a unit area over a stated time interval [Wh/m2 or kWh/m2]. Spatial grid In digital cartography the term applies to the minimum size of the grid cell or in the other words resolution minimal size of the pixels in the digital map Uncertainty Is a parameter characterizing the possible dispersion of the values attributed to an estimated irradiance/irradiation values. In this report, uncertainty assessment of the solar resource estimate is based on a detailed understanding of the achievable accuracy of the solar radiation model and its data inputs (satellite, atmospheric and other data), which is confronted by an extensive data validation experience. The second important source of uncertainty information is the understanding of quality issues of ground measuring instruments and methods, as well as the methods correlating the ground-measured and satellite-based data. In this report, the range of uncertainty assumes 80% probability of occurrence of values. Thus, the lower boundary (negative value) of uncertainty represents 90% probability of exceedance, and it is also used for calculating the P90 value. Water vapour Water in the gaseous state. Atmospheric water vapour is the absolute amount of water dissolved in air. Page 7 of 39 World Bank, Global ESMAP Initiative, Renewable Energy Resource Mapping: Solar − Maldives (Project ID: P146018) Model Validation Report – Preliminary Results 1 SUMMARY Background This Model Validation Report presents results of preliminary validation of solar resource and meteorological modelled data, within Phase 1 of the project Renewable Energy Resource Mapping for the Republic of the Maldives. This part of the project focuses on solar resource mapping and measurement services as part of a technical assistance in the renewable energy development implemented by the World Bank in Maldives. It is being undertaken in close coordination with the Ministry of Environment and Energy (MEE) of Maldives, the World Bank’s primary country counterpart for this project. This project is funded by the Energy Sector Management Assistance Program (ESMAP) and Asia Sustainable and Alternative Energy Program (ASTAE), both administered by the World Bank and supported by bilateral donors. Objective, data and methods The objective of this report is to document validation of solar resources calculated by satellite-based model SolarGIS and validation of meteorological data derived from the numerical weather model CFSR and CFSv2. Inventory in Chapter 3 identifies the existing data sources in the region: solar, aerosol and meteorological data. Aerosol data (more specifically Aerosol Optical Depth, AOD) from the MACC-II model is evaluated in Chapter 4, as this data is one of the inputs to SolarGIS clear-sky model. Chapter 5 shows relative comparison of SolarGIS GHI and DNI to other modelled databases. This chapter includes also the validation of SolarGIS in respect to solar resource measurements available in tropical climate of Asia. Chapter 6 shows validation of meteorological parameters that are delivered in the form of site-specific data sets and maps. Chapter 7 summarizes validation results in the interim estimate of uncertainty. Results Validation shows stable performance of SolarGIS model in the equatorial tropical region, though with higher uncertainty (compared to some other geographical regions). The SolarGIS uncertainty of the model can be reduced in Maldives, providing that high-quality solar resource measurements are available. The modelling results are presented in the Solar Modelling Report 129-01/2015. Page 8 of 39 World Bank, Global ESMAP Initiative, Renewable Energy Resource Mapping: Solar − Maldives (Project ID: P146018) Model Validation Report – Preliminary Results 2 MODEL QUALITY INDICATORS The performance of satellite-based models, for a given site, is characterized by the following indicators: 1. Bias characterizes systematic model deviation at a given site; 2. Root Mean Square Deviation (RMSD), Standard deviation (SD) and Mean Average Deviation (MAD), which indicate spread of error for instantaneous values (typically hourly or sub-hourly); 3. Kolmogorov-Smirnoff index (KSI) characterizes representativeness of distribution of values. This indicator is applied only for solar resource data. Focus of this report is validation and uncertainty assessment of SolarGIS solar resource data that are derived in the form of spatial and site-specific data products. Meteorological data are also validated as they are used in the site-specific times series and Typical Meteorological Year data (TMY). Air temperature is also used as a spatial data product. Only quality-controlled measurements from high-standard sensors can be used for objective validation of satellite-based solar model, as issues in the ground measured data result in skewed evaluation results. Typically, bias is considered as the first indicator of the model accuracy. While knowing bias helps to understand a possible error of longer-term estimate, other accuracy indicators should be also considered for a complete understanding of the model performance. Mean Average Deviation (MAD) and Root Mean Square Deviation (RMSD) are important for estimating the accuracy of energy simulation and operational calculations (monitoring, forecasting). Kolmogorov Smirnoff Index (KSI) reveals issues in the model’s ability to represent specific solar radiation conditions. Focusing too much on bias (systematic deviation) may lead to incorrect judgement when comparing different satellite-based models. Even if bias is similar, other accuracy indicators (RMSD, MAD and KSI) may indicate substantial differences in performance of models. Validation statistics for one site may not provide representative picture of the model performance in a given geographical conditions. The reason is that one particular site may be affected by a local microclimate or by hidden issues in the ground-measured data. Therefore, the model should be evaluated at several validation sites. Ability of the model to estimate longterm values should be evaluated analysing two measures for a set of validation points [1]: • Mean bias deviation, which indicates whether the model has overall tendency to overestimate or to underestimate the measured values. • Standard deviation of biases, which shows the range of deviation of the model estimates from ground measurements (statistically one standard deviation characterizes 68% probability of occurrence). Good satellite models are consistent in space and time, and thus the validation at several sites within one geography provides a robust indication of the model accuracy in geographically comparable regions elsewhere. Besides bias and RMSD, the ability of the model to simulate representatively sub-hourly values for all conditions (especially high and low light conditions) is very important for optimisation of the solar power plants. Two evaluation studies have been conducted independently by University of Geneva [1, 2]. Both studies analyse features of existing solar radiation models based on processing of satellite data. The studies show that SolarGIS model demonstrates robust performance in all indicators. Page 9 of 39 World Bank, Global ESMAP Initiative, Renewable Energy Resource Mapping: Solar − Maldives (Project ID: P146018) Model Validation Report – Preliminary Results 3 INVENTORY OF SOLAR ATMOSPHERIC AND METEOROLOGICAL VALIDATION DATA 3.1 Solar resource measurements Public data Solar radiation, unlike other basic meteorological parameters, is measured only at few meteorological stations in Equatorial Asia. Solar measurements are collected by various organizations: by international or regional professional networks, meteorological agencies or universities. Access to these data may be restricted by data usage polices. Inventory shows that − in geographic conditions comparable to those of Maldives − only few sources provide data with sufficient quality required for the model validation (Table 3.1). Table 3.1: Sources of solar resource validation data Network Description BSRN Baseline Surface Radiation Network provides near-continuous, long-term, in situ-observed broadband irradiances (solar and thermal infrared) and certain related parameters from a network of more than 50 globally diverse sites. Data usually include measurements of Global Horizontal Irradiance (GHI), Diffuse Horizontal Irradiance (DIF) and Direct Normal Irradiance (DNI). http://www.bsrn.awi.de/ http://www.bsrn.awi.de/en/data/data_retrieval_via_pangaea/ GAW Global Atmospheric Watch Program of World Meteorological Organization focuses on better understanding of interactions between the atmosphere, the oceans and the biosphere. It provides solar radiation data for several locations globally. http://www.wmo.int/pages/prog/arep/gaw/gaw_home_en.html http://wrdc.mgo.rssi.ru/ BoM Easy-available data are also from meteo stations that are part of national meteorological network operated by Bureau of Meteorology, Australia. ARM Atmospheric Radiation Measurement program established by U.S. Department of Energy collects data with the aim to improve understanding and representation, in models, of clouds and aerosols and their interactions and coupling with the Earth’s surface. Part of the measurement campaign is acquisition data on solar radiation. Data usually include GHI, DIF and DNI measurements. http://www.arm.gov/ MMS Maldives Meteorological Service provided sunshine hours data from the Campbell–Stokes recorder for Malé, Hulhulé airport for a period from January 2003 to December 2012. This data was used only for indicative comparison with SolarGIS DNI. http://www.meteorology.gov.mv/met/ Before using the ground-measurements for the model validation, they were quality controlled (Chapter 5.1). In general, measurements from networks such as BSRN, GAW, ARM and BoM have quality as they are based on top-accuracy measuring instruments and are well maintained. This allows using them for the model validation. It is to be noted that sunshine recorder has lower accuracy and the data from MMS provides only indicative information when compared to SolarGIS. List of public solar resource measuring stations, used for validation of SolarGIS model (Chapter 5.2) is summarized in Table 3.2. Their position is shown in Figure 3.1. Page 10 of 39 World Bank, Global ESMAP Initiative, Renewable Energy Resource Mapping: Solar − Maldives (Project ID: P146018) Model Validation Report – Preliminary Results Figure 3.1: Position of sites, for which SolarGIS validation and data comparison was performed Table 3.2: Solar measuring stations used for SolarGIS validation and data comparison Site name Country Source Latitude Longitude Altitude GHI DNI Period [°] [°] [m a.s.l.] Gan Maldives ARM -0.69048 73.150048 3 YES YES 09/2011 – 02/2012 Cocos (Keeling) (Australia) BSRN -12.189 96.8344 3 YES YES 10/2004 – 12/2008 Islands Papua New Momote BSRN -2.058 147.425 6 YES YES 01/2004 – 04/2012 Guinea Minamitorishima Japan BSRN 24.2883 153.9833 7 YES YES 04/2010 – 11/2012 Marshall Kwajalein BSRN 8.72 167.731 10 YES YES 04/2007 – 02/2010 Islands Nauru Island Nauru BSRN -0.521 166.9167 7 YES YES 01/2003 – 12/2008 Darvin airport Australia BoM -12.4239 130.8925 30 YES YES 01/1999 – 12/2010 Broome airport Australia BoM -17.9475 122.2353 7 YES YES 01/1999 – 06/2010 Bukit Indonesia GAW -0.2019 100.3181 864 YES YES 01/2000 – 03/2012 Kototabang Sunshine Malé* Maldives MMS 4.1927 73.5281 1 01/2003 – 12/2012 hours * Sunshine hours from Malé (Hulhulé airport) were used only for indicative comparison. Private initiatives A number of solar measuring stations are deployed by companies active in solar energy project development in the region. The measured data are used for commercial and technological assessment of solar resource for particular projects and they are not publically available. 3.2 Solar resource modelled data Public databases There are several modelled databases available for Maldives (Table 3.3). In general, the databases based on the interpolation of ground-measured data, such as Meteonorm [3] are less reliable in regions with sparse spatial coverage of meteorological stations. The global database NASA SSE [4] is computed by empirical models from satellite and atmospheric data with very coarse spatial resolution, which results in coarse and regionally less reliable climate patterns. SWERA/NREL database has medium spatial resolution and is computed using CSR model by NREL [5, 20], thus only showing overview perspective. The closest to SolarGIS is satellite-based Page 11 of 39 World Bank, Global ESMAP Initiative, Renewable Energy Resource Mapping: Solar − Maldives (Project ID: P146018) Model Validation Report – Preliminary Results database PVGIS CMSAF, however the data are not updated regularly and are available only as long-term averages [6]. Implementation of these databases is static, and they are not updated regularly. Table 3.3: Inventory of solar resource models for Maldives Model Data source Data spatial Parameter Time resolution Period resolution of available data NASA SSE Satellite + model 110 km x 110 km GHI, DNI Long-term monthly 1983 – 2005 Interpolation and Meteonorm 7.1 Ground + satellite GHI, DNI Long-term monthly 1981 – 2010 satellite data Long-term monthly PVGIS CMSAF MFG satellite 3 km x 3 km GHI, DNI 1999 – 2011 (hourly) SWERA/NREL Model 40 km x 40 km GHI, DNI Long-term monthly 1985 – 1991 SolarGIS* MFG satellite 3 km x 3 km GHI, DNI 30 minutes 1999 – 2015* * SolarGIS database is continuously updated on daily basis Commercial satellite-based databases There are few solar databases developed and maintained by commercial entities that provide solar radiation data to customers for a fee. Most of these databases are based on the use of satellite data, but they differ in the model implementation and use of input data (e.g. aerosols, water vapour), therefore results may differ significantly. These databases differ also in spatial coverage, spatial and temporal resolution, operational update and other parameters. Besides quality of data, important for a user is easy access, ability of the system to deliver updated data, and support by services, such as site adaptation, derived data products (e.g. TMY), bankable solar resource assessment, map services and others. To our knowledge, for Maldives, besides SolarGIS, the following commercial databases are available: SOLEMI, 3TIER and IrSOLaV. More information about these databases is available in [1, 2]. 3.3 Atmospheric data Along with the clouds, aerosols are the most influential factor controlling GHI and DNI irradiance in the region, especially during cloud-free weather situations. In general, the atmospheric turbidity is mostly influenced by aerosols in the form of burning biomass, soil particles, locally by human activities (industry, transport and urbanization) and particles transported from other regions. Geography creates specific conditions for local distribution and transport of aerosols. Combined influence of these factors results in varying atmospheric pollution both spatially and temporarily. The accurate model description of aerosols is difficult due to several factors: • Aerosols have high spatial and temporal variability, • There is insufficient number of aerosol-specialized meteo stations, and often they have only short period of measurements, • In tropical climate of islands, there are limited possibilities for detailed description of aerosol sources needed by chemical-transport atmospheric models, • Arid and semiarid conditions make it difficult to use satellite measurements of aerosols, • In general, dynamics of aerosols increases in a complex terrain (this is not a case of Maldives). For aerosol characterization a data from chemical transport model MACC-II is used in SolarGIS. The original data with resolution of ca. 85 km and 125 km is post-processed by a) regional adaptation to remove systematic regional deviation of MACC-II database and b) altitude correction to better reflect local terrain conditions (does not apply in Maldives). Understanding of nature of the modelled aerosol data helps to indirectly evaluate the satellite based SolarGIS model. For this purpose a raster data from aerosol model was compared to point aerosol measurements from AERONET [7]. It must be noted that comparison with AERONET sites is indicative only, but it helps to Page 12 of 39 World Bank, Global ESMAP Initiative, Renewable Energy Resource Mapping: Solar − Maldives (Project ID: P146018) Model Validation Report – Preliminary Results understand the correlation of point-measured data with a coarser resolution of the modelled data. Such comparison does not constitute fully independent verification of the MACC-II model outputs since the model itself uses this data on the input, as well as tend so f other data sources. Figure 3.2 shows location of AERONET stations used for atmospheric data validation. Two stations are found in Maldives: Gan and Hanimaadhoo. Fit of aerosol data to ground measurements is important indirect indicator of performance of satellite-based model (for SolarGIS this is evaluated in Chapter 4.2). Figure 3.2: Position of two AERONET stations in Maldives 3.4 Meteorological measurements The validation procedure was carried out by comparison of modelled data with ground-measured data at five sites operated, and kindly provided, by Maldives Meteorological Service (MMS). Also two other meteo sites with similar climate are evaluated (source National Climatic Data Center NCDC provided by NOAA). The position of all selected meteo sites is shown in Figure 3.3. Meteorological data for Malé (Hulhulé airport) station was provided both by MMS (period of 2 years) and also was found in the NCDC network (period of 6 years) It must be noted that time period of data comparison for the MMS network is relatively short (Table 3.4). Comparison with two other meteo stations in the region is performed for a time period 2008 to 2010 (CFSR model) and for 2011 to 2013 (CFSv2 model) [8, 9]. Page 13 of 39 World Bank, Global ESMAP Initiative, Renewable Energy Resource Mapping: Solar − Maldives (Project ID: P146018) Model Validation Report – Preliminary Results Table 3.4: Meteo stations in the region considered for validation of CFSR and CFSv2 model outputs Meteo station Data source Time period Latitude* Longitude* Elevation* [º] [º] [m a.s.l.] NSF Diego Garcia, BIOT NOAA 01/2008 – 12/2013 -7.3133 72.4111 3 Cocos, Keeling Islands, AU NOAA 01/2008 – 12/2013 -12.1830 96.8330 4 Malé, Hulhulé, MV NOAA 01/2008 – 12/2013 4.1927 73.5280 0 Hanimaadhoo, MV MMS 01/2007 – 12/2008 6.7463 73.1686 2 Malé, Hulhulé, MV MMS 01/2007 – 12/2008 4.1927 73.5280 0 Kadhdhoo, MV MMS 01/2007 – 12/2008 1.8583 73.5197 0 Kaadedhdhoo, MV MMS 01/2007 – 12/2008 0.4883 72.9961 0 Gan, MV MMS 01/2007 – 12/2008 -0.6905 73.1501 0 * Accurate geographical position and elevation of meteorological station may deviate slightly from the one in the table. Figure 3.3: Position of meteorological stations considered for validation of CFSR and CFSv2 model outputs 3.5 Meteorological models Table 3.5 gives an overview of selected modelled meteorological data available for the region. Chapter 4.2 in the Solar Modelling Report 129-01/2015 gives more insight into global meteorological models. These models are run by meteorological organisations such as US National Oceanic and Atmospheric Administration (NOAA), European Center for Medium range Weather Forecasting (ECMWF) or Canadian Meteorological Centre (CMC). Global meteorological models serve different purposes like weather forecasting, modelling long-term climate processes and helping to understand weather phenomena in a global scale. Page 14 of 39 World Bank, Global ESMAP Initiative, Renewable Energy Resource Mapping: Solar − Maldives (Project ID: P146018) Model Validation Report – Preliminary Results Table 3.5: Some meteorological models available in the region Database name Source Primary spatial Primary time Period resolution resolution CFSR NOAA 0.312° x 0.312° 1 hour 1979 to 2010 CFSv2 NOAA 0.20° x 0.20° 1 hour 2011 to present GFS NOAA 0.20° x 0.20° 3 to 6 hours 1991 to present ERA-Interim ECMWF 0.75°x 0.75° 6 hours 1979 to present GDPS MSC 0.225° x 0.225° 3 hours 2010 to present Meteonorm Ground-measurements Interpolation Long-term monthly 2000 to 2009 Accuracy of modelled meteorological data for a specific geographical location cannot compete with the accuracy of well-maintained on-site meteorological sensors. However advantages of the modelled data are numerous: they cover large territories (some are global), they are free of maintenance and calibration issues, in case of reanalysis products they also ensure seasonal and long term stability, long history, almost 100% availability and they offer data from any location on the Earth. This makes them a good choice for preliminary solar energy simulations. The Meteonorm database is also mentioned in Table 3.5. It is a different type of weather database, based on ground measurements from a number of (8325) meteorological stations, where site-specific information is calculated by interpolation of monthly averages. Monthly averages are, in the second step statistically disaggregated to synthetic hourly data representing one year. This approach has limitations due to its static character (there is no systematic update) and limited performance in areas with sparse network of meteorological stations. Although this database was historically popular, with today’s computing and modelling options, this approach is overcome. In the delivery for Maldives, the meteorological parameters are derived from CFSR and CFS v2 models. Water vapour parameter - for solar resource model - is partially derived also from GFS database. Page 15 of 39 World Bank, Global ESMAP Initiative, Renewable Energy Resource Mapping: Solar − Maldives (Project ID: P146018) Model Validation Report – Preliminary Results 4 VALIDATION OF AEROSOL DATA Along with clouds, aerosol data is one of the most important parameters as it controls accuracy of solar models especially in cloudless situations. Atmospheric aerosols represent a complex of liquid and solid particles originating from different sources, e.g. soil particles, sea salts, burning biomass, industrial and traffic pollution and pollen. Aerosols have high spatial and temporal variability and complex behaviour in terms of absorption and scattering of solar irradiance. Changing aerosol concentrations in the atmosphere can trigger variability of GHI in the range of 0% to 7%, occasionally up to 10%. In case of DNI, the variable aerosols can trigger day-by-day changes of DNI as much as 40% or even more. In solar modelling, aerosols are represented by the parameter called Aerosol Optical Depth (AOD). 4.1 Evaluation of MACC-II Aerosol Optical Depth data MACC-II aerosol data [10, 11] are used in the SolarGIS model, and they provide good representation of temporal as well as spatial variability of atmospheric load by aerosols. Despite these qualities, the data may experience systematic deviation in some regions [12]. We evaluate MACC-II AOD data using measurements from two AERONET stations [7] located in Gan and Hanimaadhoo (Figure 3.2). Data for Gan is available only for few months, as the measurements were taken during a short scientific experiment. Figure 4.1 and 4.2 demonstrate an accuracy of the MACC-II database used in the SolarGIS model. Comparison of post-processed daily MACC-II data and 15-minute AERONET ground measurements for Gan and Hanimaadhoo shows good fit. Seasonal profiles as well as short (several days) extreme situations are well represented, but in situations with very low aerosol concentration a slight overestimation of MACC-II AOD is seen. On the other hand, for some of high concentration situations, the modelled AOD may be slightly underestimated. The discrepancy visible in the plot also arises from high frequency (15-min) site-specific AERONET values that create a natural mismatch when compared with daily summaries of regionally-smoothed MACC-II data. Figure 4.1: Comparison of daily summaries from MACC-II model with 15-min AERONET data. ARM (Gan) and MCO (Hanimaadhoo) AERONET stations Page 16 of 39 World Bank, Global ESMAP Initiative, Renewable Energy Resource Mapping: Solar − Maldives (Project ID: P146018) Model Validation Report – Preliminary Results Some discrepancies and spread of values in the data for both meteo stations are attributed to limited spatial and temporal resolution of the MACC-II database, which not capable to represent with sufficient accuracy higher- frequency changes of the specific local conditions recorded in the AERONET data. Figure 4.2: Comparison of Aerosol Optical Depth for Hanimaadhoo AERONET 675 nm (x-axis) and MACC 670 nm (y-axis); blue points: original MACC data; red points: MACC data regionally-adapted by SolarGIS method To understand potential issues related to AOD in this region, the MACC-II data for the Malé site were compared with AOD data computed from other satellite missions: Terra MODIS, Aqua MODIS, Terra MISR, Envisat MERIS [13, 14]. While MACC-II model provides systematically daily values, values from satellite databases (MODIS, MERIS, TERRA MISR) have irregular time resolution, determined by the availability of cloudless days (Figure 4.3). All compared databases show the same pattern with relatively stable aerosols over the whole period. Although the range of values in different databases is similar, some differences in high values can be identified. Limited number of available ground-measured data does not allow in-depth evaluation of the accuracy of the aerosol data. Moreover, the measurements from Gan were available only for a very short period, thus this site gives only limited information. In general, analysis of available AOD data shows good representativeness of the MACC-II database. Small discrepancies could be observed and may be reduced in Phase 2 of this project, by use of high resolution and high quality local solar measurements. Page 17 of 39 World Bank, Global ESMAP Initiative, Renewable Energy Resource Mapping: Solar − Maldives (Project ID: P146018) Model Validation Report – Preliminary Results Terra MODIS Aqua MODIS Envisat MERIS Terra MISR MACC-II Figure 4.3: Aerosol Optical Depth for Malé from five different AOD databases Plots of satellite-based data were produced with by Giovanni online data system, NASA Langley ASCD [15] 4.2 Seasonal variability of Aerosol Optical Depth SolarGIS uses AOD input data, derived from the MACC-II model, for the wavelength 670 nm. The MACC-II model captures high temporal variability of aerosols, thus it reduces uncertainty of instantaneous GHI and DNI estimates. Figure 4.4 shows typical monthly variability of aerosols in central and Northern Indian Ocean. Maldives are located in a central zone with more stable and relatively low aerosol load. Northern Islands may be influenced by increased aerosol load in a period from June to September. The lowest aerosol load in the atmosphere can be observed from December to February. Islands close to the equator have lower seasonal variability. From the global perspective, Maldives is a region with relatively low aerosol load (Figure 4.5), but still having influence on the dynamics of solar resource, especially DNI. Page 18 of 39 World Bank, Global ESMAP Initiative, Renewable Energy Resource Mapping: Solar − Maldives (Project ID: P146018) Model Validation Report – Preliminary Results Figure 4.4: Monthly-averaged aerosol maps (AOD 670) derived from the MACC-II database and adapted for the SolarGIS model. Period 2003 to 2013 Figure 4.5: Average annual aerosols − Maldives in the global context AOD 670 nm is computed by the MACC-II model and adapted for SolarGIS. Period 2003 to 2012. Values are dimensionless. Page 19 of 39 World Bank, Global ESMAP Initiative, Renewable Energy Resource Mapping: Solar − Maldives (Project ID: P146018) Model Validation Report – Preliminary Results 5 VALIDATION OF SOLAR RESOURCE DATA 5.1 Quality control of solar validation data 5.1.1 Measurements by high accuracy solar sensors Prior to comparison with satellite-based solar resource data, the ground-measured solar irradiance was quality- controlled by GeoModel Solar. Quality control (QC) was based on methods defined in SERI QC procedures and Younes et al. [16, 17] and also by the GeoModel Solar’s in-house developed tests. The ground measurements were inspected also visually, mainly for identification of shading and other regular data error patterns. As an example, Figure 5.1 shows results of QC in two stations: Cocos (Keeling) Islands and Gan (Maldives). The colours indicate the following flags: • Green: data passed all tests • Grey: sun below horizon • White strips: missing data • Red and violet: GHI, DNI and DIF consistency problem or problems with physical limits • Dark grey: other issue. Figure 5.1: Quality control of data measured at Cocos Islands (top) and Gan (bottom) stations X-axis: date of measurement, Y-axis: time of measurements; colour – various QC flags (read above). Figure 5.1 shows slightly increased occurrence of issues: for short periods we have identified an inconsistency between GHI, DNI and DIF component (red colour) and measurements outside of physical limits. Data from Cocos Islands contain several periods with gaps; in the Gan station the missing data were identified only for a very short period. Page 20 of 39 World Bank, Global ESMAP Initiative, Renewable Energy Resource Mapping: Solar − Maldives (Project ID: P146018) Model Validation Report – Preliminary Results Quality control shows that solar radiation data is often affected by disturbances, which result in various errors. Among typical errors there are missing values, short periods of values with inconsistency between the solar components or values can be out of physical limits. In many stations, also shading from surrounding terrain or objects is observed. These errors were identified, to a various extent, in the data from all solar measuring stations. Availability of measurements for all three components (GHI, DNI and DIF) allows performing more complex consistency test, which reveals issues in data (such as incorrect sun tracking) that would otherwise remain hidden in the data. Very important is also visual control of the data that helps identifying systematic issues such shading, reflections or miscalibration of instruments. Quality-control procedures pre-qualified data for the model: the data readings with identified issues were flagged and excluded from further analyses. Table 5.1 summarizes percentage of excluded data by QC tests. Table 5.1: Data that did not pass through quality control [%] Type of test (numbers show percent of total volume of data, assuming 24-hour cycle) Sun below Test for Visual Consistency test Total excluded horizon physical limits test/other (GHI – DNI – DIF) data samples Gan 49.8 2.9 0.0 0.2 3.1 Cocos (Keeling) 50.4 3.4 6.3 0.0 9.7 Islands Momote 49.4 8.1 1.2 0.6 9.9 Minamitorishima 49.1 7.0 0.1 0.2 7.3 Kwajalein 49.7 5.1 0.0 0.0 5.1 Nauru Island 49.5 4.1 0.0 2.3 6.4 Darvin airport 49.8 3.7 4.2 0.0 7.9 Broome airport 50.1 2.6 3.3 0.0 5.9 Bukit Kototabang 50.0 11.1 8.5 0.0 19.6 Based on our experience from validating a large number of ground-measured data, here we propose some recommendations to consider during the Phase 2 measuring campaign: • Data to be used for the model validation must come from high accuracy instruments; technical description of instruments and information on their calibration status must be available. The equipment must be maintained and frequent cleaning of sensors must be applied. • Use of just one or two sensors (GHI, DNI), without redundant (DIF) measurements does not allow applying (very valuable) redundancy quality control algorithms. • Solar trackers and instruments should be preferably mounted about 1 m to 1.5 m above ground or roof on a stable concrete or metal platform. • Data cleaning should be systematic and logged. • Data should be quality checked on a continuous basis. Some types of logger have software which can automatically pre-flag errors. Data should be provided for processing with flags for above-mentioned problems, to avoid mistaken use of erroneous data. • Regular scheduled visits on the station every few months could prevent common issues such a tracker misalignment and issues with sensor levelling, with PV power supply or battery. At this stage of the project accurate long term solar measurements were not available for Maldives, except few months of measurements from the ARM network operating at Gan Island. There is a possibility of having access to solar radiation data measured at the Atmospheric Observatory in Hanimaadhoo Island, if successfully negotiated. 5.1.2 Measurements by Campbell-Stokes recorder Maldives Meteorological Service provided data from the Campbell–Stokes recorder. We analysed measurements for Malé, Hulhulé airport for a period of 10 complete years (January 2003 to December 2012). Page 21 of 39 World Bank, Global ESMAP Initiative, Renewable Energy Resource Mapping: Solar − Maldives (Project ID: P146018) Model Validation Report – Preliminary Results This type of instrument consists of a clear-glass sphere that focuses the sun rays onto a strip chart, producing a charred path when there is bright sunshine (Figure 5.2). Data from the sunshine recorder have its limitations, but it is worldwide acceptance as a device for measuring sunshine duration. The installations started in year 1882 st and the monitoring lasts till 21 century, ensuring a long record of heliographic data. In 1962, the Campbell- Stokes sunshine recorder was adopted by the World Meteorological Organization (WMO) as the Interim Reference Sunshine. The length of the path determines the bright sunshine duration. The lower limit for bright 2 2 sunshine (based on a Campbell-Stokes recorder) is between 70 W/m (very dry air) and 280 W/m (very humid 2 air). In 2003, the WMO established its threshold value at 120 W/m and therefore, this value has been set as the sensitivity threshold in electronic sensors. Figure 5.2: Campbell–Stokes recorder mounted at the meteo station at Malé, Hulhulé airport Climate in Maldives is tropical equatorial with high humidity (the wet season experiencing humidity levels of above 80% on average and the dryer months still as high as 75%) what has an impact on the performance of Campbell–Stokes, namely threshold value of direct irradiation (effective radiation) at which sun rays start to produce path on the paper stripe. This type of device is also very sensitive to precipitation and obtained results are often subjective (depends greatly on to the person analysing the data). Figure 5.3: Comparison of measured and modelled daily sunshine hours. Modelled data is based on processed SolarGIS Direct Normal Irradiation with 6-minute time step data (Right) and dependence of fit accuracy on the threshold value (Left). Page 22 of 39 World Bank, Global ESMAP Initiative, Renewable Energy Resource Mapping: Solar − Maldives (Project ID: P146018) Model Validation Report – Preliminary Results To compare the measurement by Campbell–Stokes with SolarGIS satellite-based data, the SolarGIS Direct Normal Irradiation for Hulhulé airport has been generated with 6-minute time step by time interpolation of the original 30-minute data (to match the temporal resolution of the provided local measurements). To calculate 2 sunshine hours from SolarGIS DNI the variable threshold was applied ranging from 190 to 250 W/m (Figure 5.3 Left). It was found that the best match (Bias close to 0h and RMSE close to 1h) occurs for the DNI value of 215 to 2 217 W/m . For that DNI value the match between measured and modelled data is very satisfactory (Figure 5.3 Right). 5.2 Validation of solar resource model 5.2.1 Comparison of SolarGIS to other models Solar irradiance for Maldives is calculated by the SolarGIS model (Chapter 3.2 of the Solar Modelling Report 128-01/2015). In this Chapter, the annual SolarGIS average is compared to five other data sources with different temporal and spatial resolution and different time coverage (Table 3.4). Six representative sites are used, as described in Chapter 6.1 of the Solar Modelling Report 128-01/2014). Comparison of the databases shows discrepancies (Tables 5.2 and 5.3), which are determined by their different characteristics: • Applied model approaches • Type and quality of input data • Time representation • Spatial and temporal resolution of the output databases. Table 5.2: Comparison of SolarGIS long-term yearly GHI average with four different models 2 Global Horizontal Irradiation [kWh/m ] Database Gan Hanimaadhoo Hulhulé Kadhdhoo NASA SSE 2129 2118 2129 2126 Meteonorm 7 1866 1922 1824 1828 PVGIS CMSAF 1855 2137 2137 2151 NREL 1918 1918 1918 1918 SolarGIS 2065 2040 2059 2062 Standard deviation of GHI annual values 6.3% 5.1% 6.8% 6.9% Schematic assessment of GHI uncertainty (80% confidence) 8.0% 6.6% 8.8% 8.8% Expected SolarGIS uncertainty (80% confidence) 6.0% 6.0% 6.0% 6.0% Table 5.3: Comparison of SolarGIS long-term yearly DNI averages with four different models 2 Direct Normal Irradiation [kWh/m ] Database Gan Hanimaadhoo Hulhulé Kadhdhoo NASA SSE 2111 2126 2126 2104 Meteonorm 7 1405 1429 1264 1338 PVGIS CMSAF 1596 1986 2002 2060 NREL 1734 1734 1734 1734 SolarGIS 1718 1541 1635 1691 Standard deviation of DNI annual values 15.1% 16.6% 19.2% 17.5% Schematic assessment of DNI uncertainty (80% confidence) 19.3% 21.3% 24.7% 22.4% Expected SolarGIS uncertainty (80% confidence) 12.0% 12.0% 12.0% 12.0% Page 23 of 39 World Bank, Global ESMAP Initiative, Renewable Energy Resource Mapping: Solar − Maldives (Project ID: P146018) Model Validation Report – Preliminary Results In general, higher uncertainties may have be expected when comparing data representing different decades due to changes in air pollution and complex climate cycles. In addition, ground observations from the last decades may have been measured with intoruments of lower accuracy and less-stringent measuring standards. Tables 5.2 and 5.3 show dispersion of yearly GHI and DNI values between five different databases, including SolarGIS. This approach is a schematic way how to assess the solar resource uncertainty. Important is to understand the risk of possible great dispersion of the estimates. For comparison we indicate a conservative uncertainty estimate for SolarGIS (more in Chapters 5.2.3 and 7.1). The modern satellite-based databases, such as SolarGIS, have high spatial and temporal resolution, they are systematically updated and quality controlled. Modern satellite-based computing approaches are considered as the mainstream source of solar resource information for energy applications - for prefeasibility studies, project optimisation, financing, and for operation and management of solar power plants. In this context, high quality ground measurements play critical role for validation and adaptation of the satellite-based models. Therefore installing solar measuring stations is planned for Phase 2 of this project. Based on the analysis of sites from Europe, North Africa and Middle East, two intercomparison studies by Pierre Ineichen from University of Geneva [1, 2] provide independent analysis of accuracy of commercial satellite- based models. SolarGIS is compared to other databases, three of them available also for Maldives: 3Tier, SOLEMI and IrSOLaV. Table 5.4: GHI quality indicators related to satellite-based solar radiation models, [2] Global Horizontal Irradiance, GHI Mean bias Standard deviation Standard deviation 2 [W/m ] [%] of biases [%] of hourly values [%] SolarGIS 3 1 2.7 17 3Tier 4 1 3.4 21 IrSOLaV 1 0 4.0 33 Table 5.5: DNI quality indicators related to satellite-based solar radiation models, [2] Direct Normal Irradiance, DNI Mean bias Standard deviation Standard deviation 2 [W/m ] [%] of biases [%] of houry values [%] SolarGIS -11 -4 5.9 35 3Tier 17 5 12.1 49 IrSOLaV -3 -1 - 54 Table 5.6: GHI quality indicators related to satellite-based solar radiation models, [1] Global Horizontal Irradiance, GHI Mean bias Standard deviation Standard deviation [%] [%] of biases [%] of hourly values [%] SolarGIS 0 0 2.1 17 SOLEMI (Aerocom aerosols) 6 2 4.8 23 IrSOLaV 2 1 4.2 24 Table 5.7: DNI quality indicators related to satellite-based solar radiation models, [1] Direct Normal Irradiance, DNI Mean bias Standard deviation Standard deviation 2 [W/m ] [%] of biases [%] of hourly values[%] SolarGIS -6 -2 5.9 34 SOLEMI (Aerocom aerosols) -40 -11 14.5 49 IrSOLaV -1 0 12.0 49 Tables 5.4 to 5.7, show selected information from both studies, and confirm that SolarGIS has very good performance in all statistical indicators − for both Global Horizontal Irradiation and Direct Normal Irradiation. Occasionally, some databases show slightly lower Mean Bias, but this indicator may hide various problems: e.g. high bias can exist in individual sites, but it may compensate when averaged in the final figure. From the user’s perspective important parameter is Standard Deviation of biases that indicates geographical stability of the model. Full comparison can be found in both studies. Page 24 of 39 World Bank, Global ESMAP Initiative, Renewable Energy Resource Mapping: Solar − Maldives (Project ID: P146018) Model Validation Report – Preliminary Results 5.2.3 Validation at sites with high-quality GHI and DNI measurements Compared to high-quality ground measurements (Tables 5.8 and 5.9) that passed through quality control (Chapter 5.1), the SolarGIS model slightly overestimates DNI in equatorial tropics and the GHI bias is oscillating around zero. At the level of individual sites, bias of the model values (systematic deviation) is found in a narrow range (typically within ±11% for yearly DNI and ±5% for yearly GHI), thus it correspond to the expected uncertainty of the SolarGIS model [18]. Slightly higher bias in Nauru Island may be result inaccurate representation of specific local microclimate by the SolarGIS model data or may indicate also problems with ground measurements. Lower values of the KSI indicator show better match of hourly cumulative distribution of values. The validation results from the Gan Island show good fit of the model to ground measurements. However, because of only short period of measurements available (approx. 5 months) the results should be considered as indicative. The data may not be representative enough to cover all seasons with diverse weather and solar irradiance patterns. Terms Bias, RMSD and KSI are explained in Glossary. Absolute values of bias are calculated for daytime hours only. Prior to the data comparison all values were harmonized into hourly time step. Number of data pairs in Tables 5.8 and 5.9 represent all valid hourly daytime data samples from which statistical measures were calculated. Table 5.8: Global Horizontal Irradiance: bias and RMSD for validation sites Site name Bias Root Mean Square Deviation KSI Data (RMSD) pairs Hourly Daily Monthly 2 [W/m ] [%] [%] [%] [%] [-] Gan 3 0.5 17.6 7.5 2.6 27 1513 Cocos (Keeling) Islands -22 -4.8 23.4 13.3 5.7 160 19924 Momote -12 -2.6 25.3 11.7 3.7 146 19600 Minamitorishima -1 -0.2 14.2 6.8 1.4 22 10155 Kwajalein 0 0.0 17.7 8.8 1.5 23 8842 Nauru Island 22 4.2 17.9 9.7 4.8 98 7204 Darvin airport 11 2.2 18.6 8.6 3.0 76 11768 Broome airport -2 -0.3 11.8 5.8 1.9 22 12207 Bukit Kototabang -1 -0.1 31.0 14.2 2.5 122 20648 Table 5.9: Direct Normal Irradiance: bias and RMSD for validation sites Site name Bias Root Mean Square Deviation KSI Data pairs (RMSD) Hourly Daily Monthly 2 [Wh/m ] [%] [%] [%] [%] [-] Gan 11 2.7 36.2 19.3 5.1 92 1497 Cocos (Keeling) Islands -15 -3.6 45.2 24.1 6.8 156 17055 Momote 12 3.3 48.7 22.9 6.3 180 18248 Minamitorishima -2 -0.4 27.2 13.4 2.0 113 9458 Kwajalein -3 -0.8 33.5 16.5 3.1 87 8628 Nauru Island 53 11.1 36.4 21.0 11.8 228 7082 Darvin airport 15 3.0 29.9 15.2 3.9 105 10968 Broome airport 14 2.2 21.3 12.9 4.9 152 11646 Bukit Kototabang 11 5.1 69.2 39.0 6.3 163 14275 Page 25 of 39 World Bank, Global ESMAP Initiative, Renewable Energy Resource Mapping: Solar − Maldives (Project ID: P146018) Model Validation Report – Preliminary Results Further information about the data and methodology and detailed analysis of uncertainty can be consulted in [18, 19]. Comparison of validation statistics computed for solar meteo sites in equatorial tropics with similar geographical conditions shows stability of the SolarGIS model outputs, and provides confidence about the estimated uncertainty of GHI and DNI. Page 26 of 39 World Bank, Global ESMAP Initiative, Renewable Energy Resource Mapping: Solar − Maldives (Project ID: P146018) Model Validation Report – Preliminary Results 6 VALIDATION OF METEOROLOGICAL DATA 6.1 Validation sites Validation was carried out by comparison of meteorological-model data with ground-measured data for five meteorological stations provided by Maldives Meteorological Services (MMS), and three meteorological stations found in the NOAA NCDC network. Comparison with the model data is performed for a time period 2008 to 2010 (CFSR model) and for 2011 to 2013 (CFSv2 model). The time period for MMS network covers 2 years only (period 2007 to 2008). For details on the applied model please refer to Chapter 4.2 in Solar Modelling Report 129-01/2015. Position of the meteorological stations is shown in Table 3.5 and Figure 3.3. 6.2 Air temperature at 2 metres Air temperature is derived from both meteorological models by postprocessing and disaggregation from the original model resolution to 1-km grid (Table 6.1). The model data represent larger area that is why they are smoothed and in the case of Maldives it represents air temperature over the surface of an open sea. Therefore the hourly model data is not capable representing exact values of the local microclimate (as measured at a meteorological station). The modelled daily and monthly averages are very close to the measured ones (mean bias), but the hourly variability (amplitude day-time vs. night-time) of the modelled data does not match the hourly variability recorded by the ground measurements. This can be observed in the table as large bias of maximum and minimum diurnal temperature (bias min and max). Parameters bias min and max are calculated as difference between average values of minimum and maximum diurnal temperature respectively. The same effect is visible for other islands (Diego Garcia and Cocos Islands). It was also found that temperature data for Hanimaadhoo station for 2007 is not consistent with 2008 data, and that is why statistics for both years are presented separately. Figure 6.1 represents data for the Hulhulé airport. It shows that while the amplitude of the measured data is huge (up to 10ºC) the amplitude of modelled data is significantly compressed (up to 4ºC). It is impossible for coarse meteorological model to represent daily variability in Maldives. The reason is that while meteo station monitors air temperature over the nearby land surface, the grid cell of the models by large majority captures air temperature over the ocean. Table 6.1: Air temperature at 2 m: accuracy indicators of the meteorological model [ºC] CFSR model (2008 to 2010)* CFSv2 model (2011 to 2013) Bias Bias Bias RMSD RMSD RMSD Bias Bias Bias RMSD RMSD RMSD mean min max hourly daily monthly mean min max hourly daily monthly Diego Garcia 0.0 1.4 -1.7 1.2 0.6 0.2 0.0 1.4 -1.7 1.3 0.6 0.0 Malé, Hulhulé -0.8 0.6 -2.3 1.5 1.0 0.8 -1.0 0.4 -2.6 1.6 1.1 -1.0 Cocos Islands 0.5 2.1 -2.5 2.1 1.2 0.8 0.4 1.9 -2.1 2.0 1.2 0.4 1 Hanimaadhoo 3.6 4.2 3.0 3.7 3.6 3.6 - - - - - - 2 Hanimaadhoo -0.1 1.6 -2.0 1.7 0.6 0.2 - - - - - - Malé, Hulhulé -0.6 0.8 -2.3 1.5 0.9 0.6 - - - - - - Kadhdhoo -0.3 1.5 -2.2 1.7 0.7 0.4 - - - - - - Kaadedhdhoo -0.5 1.6 -2.2 1.9 0.9 0.6 - - - - - - Gan -0.1 1.7 -2.0 1.7 0.7 0.2 - - - - - - * Time period for MMS network is shown in Table 3.5 1 Measurements for the year 2007 2 Measurements for the year 2008 Page 27 of 39 World Bank, Global ESMAP Initiative, Renewable Energy Resource Mapping: Solar − Maldives (Project ID: P146018) Model Validation Report – Preliminary Results Bias is expressed as a difference between the modelled and measured average values. Even that bias is low, the distribution of hourly values shows mismatch, which is determined by the low spatial resolution of the weather model. The model is not capable to describe the daily amplitude of air temperature at the islands (small piece of land in a large mass of ocean. On the other hand, the daily amplitude in the islands is relatively low. Bias min and bias max show average systematic deviation of minimum daily values (typically early morning) and maximum daily values (typically early afternoon). The same applies for the RH, WS and WD parameters discussed in Chapters below. Figure 6.1: Scatterplots of air temperature at 2 m at Hulhulé airport. Measured values (horizontal axis) and meteorological model values (vertical axis) 6.3 Relative humidity Modelled relative humidity is calculated from the specific humidity, air pressure and air temperature. Original time resolution is 1-hour. The indirect calculation of relative humidity from the meteorological models may result in higher deviation when compared to meteorological measurements. The validation results are summarized in Table 6.2. Similarly to the case of air temperature, relative humidity exhibits good match in terms of daily and monthly averages, however the diurnal variations are not represented well by hourly model data. Figure 6.2 shows the data for the Hulhulé airport. The amplitude of diurnal changes is significantly lower in comparison to the measured data. Page 28 of 39 World Bank, Global ESMAP Initiative, Renewable Energy Resource Mapping: Solar − Maldives (Project ID: P146018) Model Validation Report – Preliminary Results Table 6.2: Relative humidity: accuracy indicators of the model outputs [%] CFSR model (2008 to 2010)* CFSv2 model (2011 to 2013) Bias Bias Bias RMSD RMSD RMSD Bias Bias Bias RMSD RMSD RMSD mean min max hourly daily monthly mean min max hourly daily monthly Diego Garcia -8 1 -14 11 10 8 -10 0 -16 15 13 10 Malé, Hulhulé -4 3 -11 8 6 4 -4 2 -11 8 6 4 Cocos Islands -2 6 -10 7 5 2 -3 4 -11 8 5 4 Hanimaadhoo -6 2 -14 10 7 -6 - - - - - - Malé, Hulhulé -6 1 -12 9 7 -6 - - - - - - Kadhdhoo -5 2 -13 9 6 -5 - - - - - - Kaadedhdhoo -5 1 -13 9 7 -5 - - - - - - Gan -7 1 -14 10 8 -7 - - - - - - * Time period for MMS network is shown in Table 3.5 Figure 6.2: Scatterplots of relative humidity at 2 m at Hulhulé airport. Measured values (horizontal axis) and meteorological model values (vertical axis) 6.4 Wind speed Wind speed is calculated from the CFSR and CFSv2 models, from 10-metre wind u- and v- components with the original 1-hourly time step resolution. Comparison of the model wind speed with on-site ground measurements is summarized in Table 6.3 and Figure 6.5. The model represents regional values for 10 m height. Figure 6.3 compares wind speed from the CFSR and CFSv2 models with the measurements. Wind direction and wind speed are strongly determined by local microclimate. Maldives are flat, that is why the modelled wind speed and wind direction fit quite well the measured values. Page 29 of 39 World Bank, Global ESMAP Initiative, Renewable Energy Resource Mapping: Solar − Maldives (Project ID: P146018) Model Validation Report – Preliminary Results Table 6.3: Wind speed: accuracy indicators of the model outputs [m/s] CFSR model (2008 to 2010)* CFSv2 model (2011 to 2013) Bias Bias Bias RMSD RMSD RMSD Bias Bias Bias RMSD RMSD RMSD mean min max hourly daily monthly mean min max hourly daily monthly Diego Garcia 1.3 2.3 0.2 2.1 1.6 1.3 1.0 2.0 -0.1 1.9 1.4 1.1 Malé, Hulhulé 0.4 1.1 -0.6 1.6 1.1 0.6 0.3 1.1 -0.7 1.5 1.0 0.3 Cocos Islands 0.6 1.2 -0.4 2.9 2.4 1.1 0.0 0.9 -1.2 2.5 2.1 0.8 Hanimaadhoo 1.1 1.6 -0.8 2.9 2.4 1.6 - - - - - - Malé, Hulhulé 0.3 2.0 -3.1 3.8 3.0 0.7 - - - - - - Kadhdhoo 0.9 1.7 -1.4 2.8 2.2 1.0 - - - - - - Kaadedhdhoo 1.1 1.7 -0.6 2.7 2.2 1.2 - - - - - - Gan 0.9 1.7 -1.5 2.8 2.2 1.1 - - - - - - * Time period for MMS network is shown in Table 3.5 Figure 6.3: Duration curves of wind speed at Hulhulé airport. CFSR/CFSv2 model versus local measurements Figure 6.4: Comparison of wind direction derived from the CFSR/CFSv2 models (left) with local measurements (right) at Malé, Hulhulé meteorological station. Page 30 of 39 World Bank, Global ESMAP Initiative, Renewable Energy Resource Mapping: Solar − Maldives (Project ID: P146018) Model Validation Report – Preliminary Results Wind direction (together with wind speed) is represented by wind rose, and this parameter is strongly determined by local microclimate (Figure 6.4). The modelled wind speed and wind direction deviate from the measured values. Wind speed data for the other meteorological stations exhibit similar characteristics, with the annual bias below 1.5 m/s. Figure 6.5 Scatterplots of wind speed at 10 m at Hulhulé airport. Measured values (horizontal axis) and CFSR/CFSv2 meteorological model values (vertical axis) The reason why the model wind speed values slightly differs from the measured data is low spatial resolution of the CFSR and CFSv2 meteorological models, which represents mainly wind over the surface of open sea, while data at micro-scale may differ from the regional scale. Since meteorological model represent larger area, the highest wind speed modelled are not present in the measured data. 6.5 Precipitation Similarly to the other meteorological parameters precipitation is strongly determined by the microclimate. Not only modelling of precipitation is difficult task but also precise measurements of rainfall require systematic maintenance of rain gauges. Rain gauges have also their limitations. Attempting to collect rain data in a hurricane can be nearly impossible and unreliable (even if the equipment survives) due to wind extremes. Table 6.4 and Figure 6.6 compare the average monthly precipitation measured at meteorological stations with the modelled data. It can be noticed that meteorological model tends to underestimate precipitation but the monthly trend is very similar. In terms of PV systems the high annual precipitation in Maldives ensures low soiling losses even without periodical cleaning of the modules. Page 31 of 39 World Bank, Global ESMAP Initiative, Renewable Energy Resource Mapping: Solar − Maldives (Project ID: P146018) Model Validation Report – Preliminary Results Table 6.4: Precipitation: Average monthly sums of precipitation [mm] Precipitation [mm] Jan Feb Mar Arp May Jun Jul Aug Sep Oct Nov Dec Sum Hanimaadhoo Measured 290 11 35 168 213 196 170 121 168 318 199 417 2307 Model 155 22 87 185 238 65 129 158 149 236 113 191 1730 Malé, Hulhulé Measured 43 26 99 168 213 210 167 187 154 292 94 233 1885 Model 141 17 90 155 208 124 161 149 197 177 87 192 1697 Kadhdhoo Measured 6 73 87 105 239 228 356 138 141 234 69 159 1835 Model 64 59 69 80 225 266 272 151 244 139 66 134 1770 Kaadedhdhoo Measured 159 9 106 224 168 99 210 163 187 251 165 283 2025 Model 130 16 93 180 224 80 180 146 160 218 111 225 1763 Gan Measured 252 20 126 150 213 64 254 188 145 274 135 212 2031 Model 206 70 114 202 248 71 146 150 164 274 170 177 1992 Figure 6.6: Comparison of average monthly precipitation from the CFSR model with local measurements mean value for all stations shown in Table 6.4 Page 32 of 39 World Bank, Global ESMAP Initiative, Renewable Energy Resource Mapping: Solar − Maldives (Project ID: P146018) Model Validation Report – Preliminary Results 7 UNCERTAINTY OF THE MODEL ESTIMATES 7.1 Solar resource parameters In Maldives, the uncertainty of Direct Normal Irradiation (DNI) and Global Horizontal Irradiation (GHI) is determined by uncertainty of the SolarGIS model and ground measurements [18], more specifically: 1. Parameterization of numerical models integrated in SolarGIS for the specific data inputs and their ability to generate accurate results for various geographical conditions: • Data inputs into SolarGIS model (accuracy of satellite data, aerosols, water vapour and terrain). • Solis clear-sky model and its capability to properly characterize various states of the atmosphere • Simulation accuracy of the SolarGIS satellite model and cloud transmittance algorithms, being able to properly distinguish different types of surface, clouds, fog, vegetation, occasional flooding, etc. • Diffuse and direct decomposition models by Perez et al. 2. Uncertainty of the ground-measurements, which is determined by: • Accuracy of the instruments • Maintenance practices, including sensor cleaning, calibration • Data post-processing and quality control procedures. Statistics, such as bias and RMSD (Chapter 5.2.3) characterize accuracy of SolarGIS model in a given validation points, relative to the ground measurements. The validation results are determined by local geography and by quality and reliability of the ground-measured data. Therefore validation for one single site represents the performance of the model only in a limited geographical extent. Only if validation for several sites is available, more consistent information about the model uncertainty can be created. From the user’s perspective, the information about the model uncertainty has probabilistic nature, which is considered at different confidence levels. Tables 7.1 and 7.2 show expert estimate of the model uncertainty assumed at 80% probability of occurrence (an equivalent to 90% exceedance) of values. Table 7.2 shows that installation of measuring stations in Maldives give a good base for reducing model uncertainty. Table 7.1: Uncertainty of the preliminary SolarGIS model estimates Yearly uncertainty Monthly uncertainty Global Horizontal Irradiation (GHI) ±6% ±8% Global Tilted Irradiation (GTI) ±7% ±9% Direct Normal Irradiation (DNI) ±12% ±15% Table 7.2: Uncertainty of estimate of the yearly solar resources: ground instruments vs. SolarGIS model Best sensors and 2 1 SolarGIS data professional maintenance DNI: Rotating Shadowband Radiometer (RSR) ±3.5% ±12% DNI: First class pyrheliometer ±1% GHI: Rotating Shadowband Radiometer (RSR) ±3.5% ±6% GHI: Secondary standard pyranometer ±2% 1 Range of uncertainty depends on climate, measurements practices and post-processing 2 Depends on the geographical ability of SolarGIS model and input data to reflect the local solar climate Page 33 of 39 World Bank, Global ESMAP Initiative, Renewable Energy Resource Mapping: Solar − Maldives (Project ID: P146018) Model Validation Report – Preliminary Results 7.2 Meteorological data Accuracy of the modelled meteorological parameters stored in the SolarGIS database was assessed by comparison with ground measurements in the geographic region. Meteorological data are derived from two different numerical models covering periods from 1999 to 2010 (CFSR model) and 2011 to 2014 (CFSv2). Taking into account the results of the comparison, the uncertainty is estimated in Table. 7.3. It was found that the modelled air temperature fits quite well the measured data in terms of monthly averages but minimum and maximum values are not represented well due to small size of the islands and coarse spatial resolution of the models. Similarly to air temperature and relative humidity, the modelled wind speed and wind direction represent larger region and are geographically smoothed in a comparison to the site measurements at a meteorological station. Table 7.3: Expected uncertainty of modelled meteorological parameters in region Annual Monthly Hourly Air temperature at 2 m <1°C <1.5°C <2.5°C Relative humidity at 2 m < 8% <10% <15% Average wind speed at 10 m <1.5 m/s <1.5 m/s <3 m/s Page 34 of 39 World Bank, Global ESMAP Initiative, Renewable Energy Resource Mapping: Solar − Maldives (Project ID: P146018) Model Validation Report – Preliminary Results 8 LIST OF FIGURES Figure 3.1: Position of sites, for which SolarGIS validation and data comparison was performed ....................... 11! Figure 3.2: Position of two AERONET stations in Maldives ................................................................................. 13! Figure 3.3: Position of meteorological stations considered for validation of CFSR and CFSv2 model outputs.... 14! Figure 4.1: Comparison of daily summaries from MACC-II model with 15-min AERONET data. ......................... 16! Figure 4.2: Comparison of Aerosol Optical Depth for Hanimaadhoo.................................................................... 17! Figure 4.3: Aerosol Optical Depth for Malé from five different AOD databases ................................................... 18! Figure 4.4: Monthly-averaged aerosol maps (AOD 670) derived from the MACC-II database ............................ 19! Figure 4.5: Average annual aerosols − Maldives in the global context ................................................................ 19! Figure 5.1: Quality control of data measured at Cocos Islands (top) and Gan (bottom) stations ......................... 20! Figure 5.2: Campbell–Stokes recorder mounted at the meteo station at Malé, Hulhulé airport ........................... 22! Figure 5.3: Comparison of measured and modelled daily sunshine hours. .......................................................... 22! Figure 6.1: Scatterplots of air temperature at 2 m at Hulhulé airport. ................................................................... 28! Figure 6.2: Scatterplots of relative humidity at 2 m at Hulhulé airport. ................................................................. 29! Figure 6.3: Duration curves of wind speed at Hulhulé airport. .............................................................................. 30! Figure 6.4: Comparison of wind direction derived from the CFSR/CFSv2 models (left)....................................... 30! Figure 6.5 Scatterplots of wind speed at 10 m at Hulhulé airport. ........................................................................ 31! Figure 6.6: Comparison of average monthly precipitation from the CFSR model with local measurements ........ 32! Page 35 of 39 World Bank, Global ESMAP Initiative, Renewable Energy Resource Mapping: Solar − Maldives (Project ID: P146018) Model Validation Report – Preliminary Results 9 LIST OF TABLES Table 3.1:! Sources of solar resource validation data ........................................................................................ 10! Table 3.2:! Solar measuring stations used for SolarGIS validation and data comparison ................................. 11! Table 3.3:! Inventory of solar resource models for Maldives ............................................................................. 12! Table 3.4:! Meteo stations in the region considered for validation of CFSR and CFSv2 model outputs ........... 14! Table 3.5:! Some meteorological models available in the region....................................................................... 15! Table 5.1:! Data that did not pass through quality control [%] ........................................................................... 21! Table 5.2:! Comparison of SolarGIS long-term yearly GHI average with four different models......................... 23! Table 5.3:! Comparison of SolarGIS long-term yearly DNI averages with four different models ....................... 23! Table 5.4:! GHI quality indicators related to satellite-based solar radiation models, [2] .................................... 24! Table 5.5:! DNI quality indicators related to satellite-based solar radiation models, [2]..................................... 24! Table 5.6:! GHI quality indicators related to satellite-based solar radiation models, [1] .................................... 24! Table 5.7:! DNI quality indicators related to satellite-based solar radiation models, [1]..................................... 24! Table 5.8:! Global Horizontal Irradiance: bias and RMSD for validation sites ................................................... 25! Table 5.9:! Direct Normal Irradiance: bias and RMSD for validation sites ......................................................... 25! Table 6.1:! Air temperature at 2 m: accuracy indicators of the meteorological model [ºC] ................................ 27! Table 6.2:! Relative humidity: accuracy indicators of the model outputs [%] ..................................................... 29! Table 6.3:! Wind speed: accuracy indicators of the model outputs [m/s] ........................................................... 30! Table 6.4:! Precipitation: Average monthly sums of precipitation [mm] ............................................................. 32! Table 7.1:! Uncertainty of the preliminary SolarGIS model estimates ............................................................... 33! Table 7.2:! Uncertainty of estimate of the yearly solar resources: ground instruments vs. SolarGIS model ..... 33! Table 7.3:! Expected uncertainty of modelled meteorological parameters in region ......................................... 34! Page 36 of 39 World Bank, Global ESMAP Initiative, Renewable Energy Resource Mapping: Solar − Maldives (Project ID: P146018) Model Validation Report – Preliminary Results 10 REFERENCES [1] Ineichen P., 2014. Long Term Satellite Global, Beam and Diffuse Irradiance Validation. Energy Procedia, 48, 1586–1596. [2] Ineichen P. Five satellite products deriving beam and global irradiance validation on data from 23 ground stations, university of Geneva/IEA SHC Task 36, 2011: http://www.unige.ch/cuepe/pub/ineichen_valid-sat-2011-report.pdf [3] Meteonorm handbook, Version 6.12, Part II: Theory. Meteotest, 2010 [4] Surface meteorology and Solar Energy (SSE) release 6.0, Methodology, Version 2.4, 2009. [5] SWERA web site. Monthly and annual average global data at 40 km resolution for Africa, NREL, 2006. [6] Huld T., Müller R., Gambardella A., 2012. A new solar radiation database for estimating PV performance in Europe and Africa, Solar Energy, 86, 6, 1803-1815. [7] Aerosol Robotic Network (AERONET), NASA. http://aeronet.gsfc.nasa.gov/ [8] CFSv2 model. http://www.nco.ncep.noaa.gov/pmb/products/CFSv2/ [9] CSFR data web site http://cfs.ncep.noaa.gov/cfsr/ [10] Morcrette J., Boucher O., Jones L., Salmond D., Bechtold P., Beljaars A., Benedetti A., Bonet A., Kaiser J.W., Razinger M., Schulz M., Serrar S., Simmons A.J., Sofiev M., Suttie M., Tompkins A., Uncht A., GEMS- AER team, 2009. Aerosol analysis and forecast in the ECMWF Integrated Forecast System. Part I: Forward modelling. Journal of Geophysical Research, 114. [11] Benedictow A. et al. 2012. Validation report of the MACC reanalysis of global atmospheric composition: Period 2003-2010, MACC-II Deliverable D_83.1. [12] Cebecauer T., Šúri M., 2012. Correction of Satellite-Derived DNI Time Series Using Locally-Resolved Aerosol Data.. Proceedings of the SolarPACES Conference, Marrakech, Morocco, September 2012. [13] NASA Multiangle Imaging Spectroradiometer (MISR). https://www-misr.jpl.nasa.gov/ [14] NASA Moderate Resolution Imaging Spectroradiometer (MODIS). http://modis.gsfc.nasa.gov/ [15] NASA Goddard Earth Sciences, Data and Information Services Center (GES DISC), Giovanni - Interactive Visualization and Analysis. http://disc.sci.gsfc.nasa.gov/giovanni [16] NREL, 1993. User’s Manual for SERI_QC Software-Assessing the Quality of Solar Radiation Data. NREL/TP-463-5608. Golden, CO: National Renewable Energy Laboratory. [17] Younes S., Claywell R. and Munner T, 2005. Quality control of solar radiation data: Present status and proposed new approaches. Solar Energy 30, 1533-1549. [18] Šúri M., Cebecauer T., 2014. Satellite-based solar resource data: Model validation statistics versus user’s uncertainty. ASES SOLAR 2014 Conference, San Francisco, 7-9 July 2014. [19] Perez R., Cebecauer T., Suri M., 2014. Semi-Empirical Satellite Models. In Kleissl J. (ed.) Solar Energy Forecasting and Resource Assessment. Academic press. [20] Renné D., George R., Marion B., Heimiller D., Gueymard Ch., 2003. Solar Resource Assessment for Sri Lanka and Maldives. NREL Report, NREL/TP-710-34645. http://www.nrel.gov/docs/fy03osti/34645.pdf Page 37 of 39 World Bank, Global ESMAP Initiative, Renewable Energy Resource Mapping: Solar − Maldives (Project ID: P146018) Model Validation Report – Preliminary Results 11 BACKGROUND ON GEOMODEL SOLAR Primary business of GeoModel Solar is in providing support to the site qualification, planning, financing and operation of solar energy systems. We are committed to increase efficiency and reliability of solar technology by expert consultancy and access to our databases and customer-oriented services. The Company builds on 25 years of expertise in geoinformatics and environmental modelling, and 14 years in solar energy and photovoltaics. We strive for development and operation of new generation high-resolution quality-assessed global databases with focus on solar resource and energy-related weather parameters. We are developing simulation, management and control tools, map products, and services for fast access to high quality information needed for system planning, performance assessment, forecasting and management of distributed power generation. Members of the team have long-term experience in R&D and are active in the activities of International Energy Agency, Solar Heating and Cooling Program, Task 46 Solar Resource Assessment and Forecasting. ® GeoModel Solar operates a set of online services, integrated within SolarGIS information system, which includes data, maps, software, and geoinformation services for solar energy. http://geomodelsolar.eu http://solargis.info GeoModel Solar is ISO 9001:2008 certified company for quality management since 2011. Page 38 of 39