(UNDER CONSTRUCTION)
manuals: https://energyRt.github.io/merra2ools/
dataset: https://doi.org/10.5061/dryad.v41ns1rtt
Overview
merra2ools package offers a set of tools and MERRA2 data subset to evaluate hourly output potential of solar and wind energy sources, as well as precipitations for weatherdependent hydro power output. The goal of the project is to provide both  the dataset and algorithms to estimate potential output and socalled capacity factors for variable energy sources, used as an input data in energy system modeling and broader application. To keep the size of the database lower than 300Gb for online publication, the original subset of MERRA2 time series have been minimally processed, rounded, and saved as scaled integers in highly compressed format provided by fst
package.
The merra2ools dataset has 41 years (19802000) of the hourly timeseries:

UTC  date and time (key) in Coordinated Universal Time (UTC) timezone;

locid  location IDs (key), an index of locations in MERRA2 dataset, from 1 to 207936;

W10M  10meter wind speed (calculated
sqrt(V10M^2 + U10M^2)
where V10M
and U10M
are northward and eastward wind at 10meter, m/s, rounded to the first decimal place);

W50M  50meter wind speed (calculated
sqrt(V50M^2 + U50M^2)
where V50M
and U50M
are northward and eastward wind at 50meter, m/s, rounded to the first decimal place);

WDIR  Direction of wind at 50meter height (calculated
atan2(V50M/U50M)
, rounded to tens);

T10M  10meter air temperature (Celsius, rounded to the nearest integer);

SWGDN  Incident shortwave land (W/m^2, rounded to the nearest integer);

ALBEDO  Surface albedo (index 0, 1, rounded to second decimal place);

PRECTOTCORR  Bias corrected total precipitation (kg/m^2/hour, rounded to the first decimal place);

RHOA  Air density at surface (kg/m^2, rounded to second decimal place).
Representation of wind with three variables (W10M
, W50M
, WDIR
) is the main difference from the original MERRA2 data (V10M
, U10M
, V50M
, U50M
). This conversion reduces the size of the database and further computational burden of wind power capacity factors.
All variables are hourly averages, UTCtime is given for a middle of every hour.
The merra2ools package includes:
 MERRA2 grid information and functions to match geolocations with MERRA2 grid;
 functions to read and subset data from the compressed files (
fst
format);
 functions and methods to evaluate solar photovoltaics hourly output/capacity factors;
 functions and methods wind speed extrapolation for higher altitudes (50200+ meters) and estimate wind power capacity factors;
 functions to fetch data from NREL’s PVWatts model and the dataset by locations (with a goal of validation the POA model and the data);
 functions and methods for “quick” figures and animated gif figures for instant evaluation of the data and the used methodology;
 long term summary statistics of the provided time series for purpose of clustering and aggregation in further modeling and analytics.
Solar Power
The package reproduces basic algorithms of solar geometry, irradiance decomposition, and the PlaneOfArray models for different types of solar PV trackers.
Wind power