Dynamic near-surface air-temperature lapse rates for the Cascade–Sierra study domain derived from MERRA-2 (2000–2020)

Description This dataset contains dynamic near-surface air-temperature lapse-rate fields for the Cascade–Sierra study domain, derived from MERRA-2 2-m air temperature (T2M), terrain elevation, and land-mask data. The dataset was generated using the dynamic lapse-rate methodology described in Whitney et al. (2025). The resulting lapse-rate fields were used to downscale meteorological forcings within the NASA Land Information System (LIS) modeling framework described in Whitney et al. (2025). The software used to generate this dataset is available as part of the NASA Land Information System Framework (LISF) repository. In this context, dynamic lapse rates refer to lapse rates that are computed directly from the spatial relationship between temperature and elevation at each time step and grid cell, rather than prescribing a constant value. At each land grid cell, lapse rates are estimated using a linear regression of temperature differences versus elevation differences within a local 3 × 3 neighborhood. The method follows the framework introduced by Rouf et al. (2020; R2020) and uses the modified R2020 configuration described in Whitney et al. (2025), including the following modifications: For grids with fewer than five land neighbors (>50% water coverage), a static lapse rate of −6.5 °C km⁻¹ is applied. Dynamic lapse rates are capped at ±10 °C km⁻¹. The dataset spans 1 June 2000 through 31 December 2020 and is provided as daily NetCDF files. Each file contains all hourly lapse-rate estimates for a single day; no temporal averaging is applied. Dataset contents This deposit includes: Dynamic lapse-rate fields for the Cascade–Sierra study domain (34.495–49.505°N, 124.69–117.50°W) One NetCDF file per day for 2000-06-01 through 2020-12-31 Files grouped into yearly ZIP archives (e.g., 2000.zip, 2001.zip, ..., 2020.zip) Variable: lapse_rate Units: K/km Each NetCDF file includes metadata describing the source dataset, source variable, input description, and method reference. Data sources The dataset is derived from: MERRA-2 reanalysis (tavg1_2d_slv_Nx) Variable: T2M (2-m air temperature) MERRA-2 data are available from: https://disc.gsfc.nasa.gov/datasets?project=MERRA-2 Static elevation and land-mask inputs were taken from MERRA-2 fields on the same grid as the temperature data. Study domain The dataset corresponds to the Sierra Nevada and Cascade Range study domain used in Whitney et al. (2025), defined approximately by 34.495–49.505°N latitude and 124.69–117.50°W longitude. Dynamic lapse rates were estimated from MERRA-2 air temperature and elevation fields for the study domain following the methodology described in Whitney et al. (2025). File format and structure Format: NetCDF Primary variable: lapse_rate Dimensions: (time, lat, lon) Units: K/km Output filenames follow the convention: MERRA2.lapse_rate.hourly.YYYYMMDD.nc where YYYYMMDD represents the calendar date (e.g., 20000601 for 1 June 2000). The filename includes hourly because each daily file retains the original hourly time steps. Although files are written as one file per day, each file contains the full set of hourly lapse-rate estimates for that day. The Zenodo deposit is organized into yearly ZIP archives, each containing the daily NetCDF files for a given year. Software and reproducibility The software used to generate this dataset is available in the NASA Land Information System Framework (LISF) repository under: https://github.com/NASA-LIS/LISF/tree/eis-freshwater2/lis/utils/nldas-3_forcing/dynamic_lapse_rates The repository includes the implementation of the dynamic lapse-rate methodology, documentation of the supported workflows, and an example workflow reproducing the Cascade–Sierra dataset archived in this Zenodo record. The methodology is described in detail in Whitney et al. (2025), which also documents its application within the NASA Land Information System (LIS) modeling framework. Related publication Whitney, K. M., et al., 2025: Quantifying the impacts of dynamic lapse regimes on snow simulations. Journal of Hydrometeorology, 26(10), 1525–1560. https://doi.org/10.1175/JHM-D-25-0021.1