Meteorological Conditions
Meteorological conditions are required as forcing fields for nearly all land surface processes to be simulated in GMELT. At present, there is considerable uncertainty in historical climate conditions of the HMA due to a number of factors: the lack of in situ observations, especially at high elevations, that can be assimilated into climate reanalysis products; the uncertainty in available weather station observations due to challenges in measuring solid precipitation; the complexity of the climate system in the region due to steep topography and associated orographic effects, and challenges in predicting the Asian monsoon season; the interference of anthropogenic factors such as black carbon and aerosols which are poorly measured and have unknown impacts on the climate system; and the sparseness of some available satellite observations, in part due to the cloudiness of the region. The highly variable land surface conditions of HMA, caused by extreme topographic variations, means that many LSM modules will require exceptionally high spatial and temporal resolution meteorological data if they are to produce meaningful output. Standard meteorological variables (air temperature, precipitation, wind speed and direction, short- and longwave radiation, humidity and pressure) will be necessary for many of the cryospheric modules, especially those that calculate the full surface energy balance. Altitudinal variations of these data will be needed to distribute observation across the landscape, and to assess mass and energy fluxes through the atmosphere. In addition, there is an increased abundance of dust and black carbon in the HMA region which has significant impacts on many land surface processes, for example through altering the albedo of snow. Datasets and models predicting the fate of these atmospheric constituents through space and time will be required to generate accurate GMELT estimates.
There are numerous approaches to generating climate products for the HMA region, and each of these approaches are represented across various HiMAT teams. Statistical models interpolate ground observations to regular grids and often include empirical adjustments to simulate physical processes, such as orographic effects. Reanalysis provides a snapshot of historical climate conditions by combining observations and numerical models through the process of data assimilation. Numerical weather prediction systems combine a data assimilation system with a weather forecast model to enable a multiple implementations of a wide range of meteorological investigations. Finally there are General Circulation Models (GCMs) that account for all the coupled atmosphere, land and ocean processes on the globe to estimate historical and predict future global climate conditions. GCMs require meteorological data as initial conditions but generally do not assimilate observations. It is important to note that many studies combine climate observations and models in complex ways. For example, often a climate reanalysis product is used to proscribe initial conditions for a GCM run, or to set the lateral boundary conditions on a WRF simulation for a specific region.