Data assimilation consists in optimally combining observations of a system with
outputs of a numerical model. In the traditional approaches, observations are
are considered to be eulerian and scalar. Recently, in the field
of earth and ocean observations, there is an increasing number of observations
that are fundamentally non-eulerian and/or non scalars:
Drifters and floats are advected by currents and bring a Lagrangian observation through their successive positions,
satellite data are measuring images that contains structured patterns.

Urbanization in the Arctic and sub-Arctic is an increasingly important anthropogenic influence on climate, and has significantly affected terrestrial ecosystems. One of the most evident effects associated with urbanization is the urban heat island (UHI), when urban and suburban areas are warmer then rural areas. The effect is more pronounced in the high latitudes.

There is evidence that the global water cycle has been undergoing an intensification over several decades as a response to increasing atmospheric temperatures, particularly in regions with skewed evaporation – precipitation (E-P) patterns such as the oceanic subtropical gyres. Moreover, observational data (rain gauges, etc.) are quite sparse over such areas due to the inaccessibility of open ocean regions. We analyzed spatial and temporal salinity trends in five subtropical gyre regions over the past six decades using Simple Ocean Data Assimilation (SODA) reanalysis.

The forecasting community has long understood the impact of dynamic instability on the uncertainty of predictions in physical systems and this has led to innovative filtering design to take advantage of the knowledge of process models. The advantages of this combined approach to filtering, including both a dynamic and statistical understanding, have included dimensional reductions and robust feature selection in the observational design of filters.