Abstract
Long time series of accurate Sea Surface Temperatures (SSTs) are needed to resolve subtle signals that may be indicative of a changing climate. Motivated by the stringent requirements on SST accuracy required for Climate Data Records (CDR) we quantify sampling errors in satellite SSTs. Infrared (IR) sensors, including the Moderate Resolution Imaging Spectroradiometer (MODIS), have sampling errors caused by incomplete coverage primarily due to clouds and inter-swath gaps (gaps between successive swaths). Unlike retrieval errors, the sampling errors are introduced when calculating mean values and in generating gap-free SST fields. This dissertation is focused on quantifying and parameterizing the global MODIS sampling errors. The MODIS-sampled SST field is generated by superimposing MODIS cloud masks on top of the Multi-scale Ultrahigh Resolution (MUR) SST field for the same day. Based on the MODIS-sampled fields, sampling errors are calculated at different temporal and spatial resolutions to examine the impacts at different scales. In order to assess the robustness of the quantification using a reasonable reference field, we compare the sampling errors quantified using MUR and those generated from another very different reference SST field— HYCOM (HYbrid Coordinate Ocean Model) Global 1/12° reanalysis. Also, sampling errors are compared for variations between El Niño and La Niña events. The climatological component of the sampling errors are calculated and assessed for its importance on sampling error estimation. Based on the error characterizations, an empirical model is proposed to parameterize the sampling errors using cloud masks and climatological or reference SST standard deviations. Global sampling errors generated from both MUR and HYCOM reference fields are significant, more so in the high latitudes, especially the Arctic. The 30°N-30°S zonal band is found to have the smallest errors; a notable exception is the persistent negative errors found in the Tropical Instability Wave (TIW) area, where the mesoscale ocean-atmosphere interaction leads to a more frequently satellite sampling above the cold sections of the wave area. The global mean sampling error is generally positive and increases approximately exponentially with missing data fraction at a fixed averaging interval, while error variability is mainly controlled by SST variability. Areas with persistent cloud cover have large sampling errors in temporally averaged SSTs. As opposed to the fact that HYCOM and MUR SSTs are substantially different globally, geophysical patterns of the sampling errors generated from HYCOM reanalysis SSTs repeat those from MUR, giving rise to sampling error differences commonly within ±0.1 K. This result support the robustness of using a reasonable reference SST field for the quantification of MODIS sampling errors, and provide the evidence of the sampling error estimates being the consequences from missing observations and not the choice of the reference field. The negative errors in the TIW area change proportionally with the SST gradient, which is recognized of being modulated by the El Niño and La Niña events. The climatology component is demonstrated to be the dominant component in the sampling errors, especially for the errors caused by spatial averaging, therefore can be a reasonable estimates for sampling errors due to spatial averaging. Yet for the sampling errors caused by temporal averaging, only 10% of the sampling errors can be attributed to the seasonal variation embedded in the climatology. We propose an empirical sampling error model by incorporating the sampling error nonlinear dependence on cloud and SST variabilities. As a result, the sampling error estimates in many regions, particularly where warm sampling errors prevail, are largely improved comparared to using only the climatological component. This dissertation initiates the global characterization and parameterization of IR sampling errors due to clouds and inter-swath gaps. The results indicate the MODIS SST sampling errors can be an important or even dominant component of the error budget of mean and gap-free SST fields. Therefore, climate data generation and interpretation of satellite-derived SST CDRs and their application must be conducted with due regard to the sampling error.