Measuring very low frequency sea level variations using satellite altimeter data

Sea-level variability in the South China Sea was investigated based on satellite altimetry, tide-gauge data, and temperature and salinity climatology. The altimetric sea-level results clearly reveal three distinct amphidromes associated with the annual cycle. The annual sea level is higher in fall/winter in the coast and shelf region and lower in summer/fall in the central sea, agreeing well with independent tide-gauge data. Averaged over the deep basin (bottom depth > 2,000 m), the annual cycle can be approximately accounted for by the steric height relative to 700 db. Significant interannual sea-level change is observed from altimetry and tide-gauge data. The interannual and longer-term sea-level variability in the altimetric data is negatively correlated (significant at the 95% confidence level) with the El Nino - Southern Oscillation (ENSO), attributed in part to the steric height change. The altimetric sea-level rise rate is 1.0 cm/year for the period from 1993 to 2001, which is consistent with the rate derived from coastal tide-gauge data and approximately accountable for by the steric height calculated relative to 700 db. The tide-gauge sea-level (steric height) rise rate of 1.05 (0.9) cm/year from 1993 to 2001 is much larger than that of 0.22 (0.12) cm/year for the period from 1979 to 2001, implying the sensitivity to the length of data as a result of the decadal variability. Potential roles of the ENSO in the interannual and longer-term sea-level variability are discussed in terms of regional manifestations such as the ocean temperature and salinity.

Dynamical downscaling employing high-resolution ocean models is widely considered as an efficient approach for modelling of regional ocean dynamics and sea-level changes using output of original coarse-resolution global general circulation models (GCMs). In this study, the historical dynamic variability and trends of sea level in the South China Sea (SCS) and the Southeast Asia Seas (SEAS) are investigated using the high-resolution regional ocean model (NEMO). Two hindcast ocean modelling experiments are conducted for the period 1960-2014. One is driven by global reanalysis data (ERA5 and ORAS5) forcings at the lateral and surface boundaries. The other is driven by global modelling oceanic data (EC-EARTH3) at the lateral boundary and by WRF-based downscaled atmospheric fields from the same parent model (EC-Earth3) at the surface boundary. Using the hindcast model runs, variability and trends of low-frequency sea-levels, as well as the driving mechanisms and the related processes are discussed, and the model performance and biases are analysed.This Research is supported by Singapore’s National Research Foundation and National Environment Agency under the National Sea Level Programme Funding Initiative (Award No. USS-IF-2020-4).

A review of sea level observations and low frequency sea-level variability in South Atlantic

For a dynamical interpretation of sea level records, estimates are needed of the isostatic, or so-called inverted barometer, signals (ηib) associated with the ocean response to atmospheric loading. Seasonal and longer-period ηib signals are evaluated over the global ocean for the period 1958–2000 using monthly sea level pressure fields from two different atmospheric reanalyses. Variability and linear trends in ηib agree well for the two reanalyses in most regions but less so over the Southern Ocean, where uncertainties in ηib seem to be largest. The standard deviation of ηib ranges from <1 cm in equatorial regions to >7 cm in the regions of the Aleutian and Iceland lows and parts of the Southern and Arctic Oceans. When compared to a global tide gauge dataset, both seasonal and interannual ηib signals are found to contribute importantly to the sea level variance in many mid- and high-latitude records, with seasonal signals important as well in tropical records from India and Southeast Asia. For these records, subtracting ηib from the data can lead to changes in variance of 40% or more. Over the period of study, linear trends in ηib are mostly negative at low and midlatitudes and can cause negative biases in tide gauge estimates of global mean sea level rise that are comparable in magnitude to the effects of postglacial rebound. In agreement with previous findings, ηib signals are found to introduce anomalous behavior in local records (e.g., substantially weaker upward trends in the Mediterranean), and their removal can also reduce formal trend uncertainties. Accounting for ηib effects can be even more important when analyzing relatively short (decadal) records, such as those available from satellite altimetry.

Continental groundwater loss during the past century has elevated sea level by up to ~25 mm. The mass unloading associated with this depletion locally uplifts Earth's solid surface and depresses the geoid, leading to slower relative sea level rise near areas of significant groundwater loss. We computed spatial variations in sea level using a model of the solid Earth's response to estimates of groundwater depletion during the past century and find large negative deviations of ~0.4 mm/yr along the coastlines of western North America and southern Asia. This approximately corresponds to the difference between rates of sea level rise measured by tide gauges in these regions since 1930 and average rates inferred from global reconstructions. Groundwater‐induced regional variations in sea level can be larger than those due to postglacial rebound and interseismic deformation and should become increasingly important in the future as both groundwater depletion and sea level rise accelerate.

Abstract. Sea-level variations in coastal areas can differ significantly from those in the nearby open ocean. Monitoring coastal sea-level variations is therefore crucial to understand how climate variability can affect the densely populated coastal regions of the globe. In this paper, we study the sea-level variability along the coast of Norway by means of in situ records, satellite altimetry data, and a network of eight hydrographic stations over a period spanning 16 years (from 2003 to 2018). At first, we evaluate the performance of the ALES-reprocessed coastal altimetry dataset (1 Hz posting rate) by comparing it with the sea-level anomaly from tide gauges over a range of timescales, which include the long-term trend, the annual cycle, and the detrended and deseasoned sea-level anomaly. We find that coastal altimetry and conventional altimetry products perform similarly along the Norwegian coast. However, the agreement with tide gauges in terms of trends is on average 6 % better when we use the ALES coastal altimetry data. We later assess the steric contribution to the sea level along the Norwegian coast. While longer time series are necessary to evaluate the steric contribution to the sea-level trends, we find that the sea-level annual cycle is more affected by variations in temperature than in salinity and that both temperature and salinity give a comparable contribution to the detrended and deseasoned sea-level variability along the entire Norwegian coast. A conclusion from our study is that coastal regions poorly covered by tide gauges can benefit from our satellite-based approach to study and monitor sea-level change and variability.

Sea level rise (over the globe as well as over the Indian Ocean) is one of the most serious issues considering its potential threat to the low-lying coastal areas and coastal population. TOPEX/Poseidon altimeter observations suggest a sea level rise over the Arabian Sea and Bay of Bengal. The rate of sea level rise over the Arabian Sea is about 0.5–3 mm/year, whereas over the Bay of Bengal, it is 0.75–6 mm/year. Major contributors to these changes in the Indian Ocean are steric effect and short-term climate variability such as El Nino–Southern Oscillation (ENSO) and the Indian Ocean Dipole. This affirms that sea level trends over north Indian Ocean get modulated by inter-annual and decadal scale natural climate variability. The inter-annual variability is stronger than decadal variability, which in turn is stronger than the long-term sea level trend. Sea level change in the Indian Ocean is about 1.5 mm/year in the past sixty years or so, whereas the global sea level trends are a bit higher. This picture completely changed in the recent years with very significant sea level rise in the Indian Ocean and is comparable to the global trends. A future projection for the sea level rise over the Indian Ocean based on the highest emission scenario is 40–67 cm for the averaged span of 2071–2100 with the base period of 1921–1990.

The rate of sea level rise (SLR) along the Southeast Coast of the U.S. increased significantly after 2010. While anthropogenic radiative forcing causes an acceleration of global mean SLR, regional changes in the rate of SLR are strongly influenced by internal variability. Here we use observations and climate models to show that the rapid increase in the rate of SLR along the U.S. Southeast Coast after 2010 is due in part to multidecadal buoyancy-driven Atlantic meridional overturning circulation (AMOC) variations, along with heat transport convergence from wind-driven ocean circulation changes. We show that an initialized decadal prediction system can provide skillful regional SLR predictions induced by AMOC variations 5 years in advance, while wind-driven sea level variations are predictable 2 years in advance. Our results suggest that the rate of coastal SLR and its associated flooding risk along the U.S. southeastern seaboard are potentially predictable on multiyear timescales.

Sea-level rise enhances carbon accumulation in United States tidal wetlands

Understanding the forcing of regional sea level variability is crucial as many people all over the world live along the coasts and are endangered by the sea level rise. The adding of fresh water into the oceans due to melting of the Earth’s land ice together with thermosteric changes has led to a rise of the global mean sea level (GMSL) with an accelerating rate during the twentieth century, and has now reached a mean rate of 3.7 mm per year according to IPCCs latest report. However, this change varies spatially and the dynamics behind what forces sea level variability on a regional to local scale is still less known, thus making it hard for decision makers to mitigate and adapt with appropriate strategies.Here we present a novel approach using machine learning (ML) to identify the dynamics and determine the most prominent drivers forcing coastal sea level variability. We use a recurrent neural network called Long Short-Term Memory (LSTM) with the advantage of learning data in sequences and thus capable of storing some memory from previous timesteps, which is beneficial when dealing with time series. To train the model we use hourly ERA5 10-m wind, mean sea level pressure (MSLP), sea surface temperature (SST), evaporation and  precipitation data between 2009-2017 in the North Sea region. To reduce the dimensionality of the data but still preserve maximal information we conduct principal component analysis (PCA) after removing the climatology which are calculated by hourly means over the years. Depending on the explained variance of the PCs for each driver, 2-4 PCs are chosen and cross-correlated to eliminate collinearity, which could affect the model results. Before being used in the ML model the final preprocessed data are normalized by min-max scaling to optimize the learning. The target data in the model are hourly in-situ sea level observations from West-Terschelling in the Netherlands. Using in-situ observations compared to altimeter data enhances the ability of making good predictions in coastal zones as altimeter data has a tendency to degrade along the coasts. The sea level time series is preprocessed by tidal removal and de-seasoned by subtracting the hourly means. To determine which drivers are most prominent for the sea surface variability in our location, we mute one driver at a time in the training of the network and evaluate the eventual improvement or deterioration of the predictions.Our results show that the zonal wind is the most prominent forcing of sea level variability in our location, followed by meridional wind and MSLP. While the SST greatly affects the GMSL, SST seems to have little to no effect on local sea level variability compared to other drivers. This approach shows great potential and can easily be applied to any coastal zone and is thus very useful for a broad body of decision makers all over the world. Identifying the cause of local sea level variability will also enable the ability of producing better models for future predictions, which is of great importance and interest.

<strong class="journal-contentHeaderColor">Abstract.</strong> While a global acceleration of sea-level rise (SLR) during the 20th century is now established, locally acceleration is more difficult to detect because additional processes play a role which sometimes mask the acceleration. Here we study the rate of SLR along the coast of the Netherlands from six tide gauge records, covering the period 1890&ndash;2000. We focus on the influence of the wind field and the nodal tide variations on the local sea-level trend. We use four generalised additive models, including different predictive variables, and a parametric bootstrap method to compute the sea-level trend. From the sea-level trend, we obtain the continuous evolution of the rate of SLR and its uncertainty over the observational period through differentiation. Accounting for the nodal cycle only or both the nodal cycle and the wind influence on sea level reduces the standard error in the estimation of the rate of SLR. Moreover, accounting for both the nodal and wind influence changes the estimated rate of SLR, unmasking an acceleration of SLR that started in the 1960s. Our best-fitting statistical model yields a rate of SLR of about 1.8 [1.4&ndash;2.3] mm/yr in 1900&ndash;1919 and 1.5 [1.1&ndash;1.8] mm/yr in 1940&ndash;1959 compared to 3.0 [2.4&ndash;3.5] mm/yr over 2000&ndash;2019. If, apart from tidal, wind effects and fluctuations, sea level would have increased at a constant rate, then the probability (the p-value) of finding a rate difference between 1940&ndash;1959 and 2000&ndash;2019 of at least our estimate is smaller than 1 %. Our findings can be interpreted as an unequivocal sign of the acceleration of current SLR along the Dutch coast since the 1960s. This aligns with global SLR observations and expectations based on a physical understanding of SLR related to global warming. A small but significant part of the long-term sea-level trend is due to wind forcing related to a strengthening and northward shift of the jet stream. Additionally, we detect a multidecadal mode of sea-level variability forced by the wind with an amplitude of around 1 cm. We argue that it is related to multi-decadal sea surface temperature variations in the North Atlantic, similar to the Atlantic Multidecadal Variability.

Projections of global sea-level rise into the future have become more pessimistic over the past five years or so. A global rise by more than one metre by the year 2100 is now widely accepted as a serious possibility if greenhouse gas emissions continue unabated. That is witnessed by the scientific assessments that were made since the last IPCC report was published in 2007. The Delta Commission of the Dutch government projected up to 1.10 m as a 'high-end' scenario (Vellinga et al 2009). The Scientific Committee on Antarctic Research (SCAR) projected up to 1.40 m (Scientific Committee on Antarctic Research 2009), and the Arctic Monitoring and Assessment Programme (AMAP) gives a range of 0.90–1.60 m in its 2011 report (Arctic Monitoring and Assessment Programme 2011). And recently the US Army Corps of Engineers recommends using a 'low', an 'intermediate' and a 'high' scenario for global sea-level rise when planning civil works programmes, with the high one corresponding to a 1.50 m rise by 2100 (US Army Corps of Engineers 2011).

Variations of the 10‐day‐averaged sea level extracted from TOPEX/POSEIDON (T/P) altimeter data are examined from October 1992 to February 1995. Data are corrected from instrumental drifts, and recent precise orbit and ocean tide corrections are used. When data from the less accurate first eight cycles are excluded and atmospheric pressure effects are not corrected, the 10‐day mean sea level varies by 5.0 mm rms. Its variations can be described by a 4.1±1 mm/yr drift, a small annual signal of 2.7 mm amplitude and a residual signal of 3.5 mm rms amplitude, and Gaussian statistics close to formal error estimates. The mean sea level rise is difficult to explain by drifts of geophysical corrections, though the case for the wet tropospheric correction and that for the orbit are uncertain. Pressure effects should be corrected relative to the global pressure averaged over the ocean. The latter shows a seasonal signal of 1 cm amplitude and apparent fluctuations of 30‐ to 70‐day periods. These fluctuations could result from errors in the atmospheric pressure fields, but they seem to be mostly related to large regional pressure evolutions. Evidence is also found for a sampling effect by the T/P orbit of pressure signals at shorter than 10‐day periods. Pressure effects on sea level are then assumed to be an inverse barometer relative to 10‐day global atmospheric pressure over the ocean, which allows analysis of the geographical repartition of the mean sea level evolution due to other effects. The mean sea level rise is comparable in the northern (4.2±2.5 mm/yr) and southern (3.7±1.9 mm/yr) hemispheres. It seems mostly related to interannual variations of the tropical ocean (41% of the global signal) and of the northern hemisphere subtropical gyres (30% of the global signal); residual noise on these regional values seems inversely proportional to the square root of the number of data points in each area. The hemispheric seasonal pressure variation relative to the globally averaged pressure over the ocean is less than 3 mm amplitude; in our calculation it is implicitely assumed that it induces a small interhemispheric water mass exchange. The amplitudes of the seasonal hemispheric sea level variations are 27 and 18 mm in the northern and southern hemispheres, respectively. These amplitudes are very close to being inversely proportional to the ocean surface in each hemisphere. The maxima of sea level in these hemispheres are reached in the last weeks of September and March, respectively. At subpolar latitudes, semiannual components are observed in each hemisphere, which are consistent with those of sea surface temperature.

First, we would like to thank Parker for his discussion and comments on our work. Parker’s main concern [1] is that the sea level is oscillating with important multi-decadal periodicities but absolutely not positively accelerating. He also argued that the global mean sea level (GMSL) acceleration computed by Jin et al. [2] seemed too large. His similar concern had also appeared in the comment of local sea level rise on Long Island, New York [3]. The most important difference between these two commented papers is that the former computes the sea level acceleration globally, while the latter locally. Parker quoted both, individual tide gauges and the time series of GMSL by satellite radar altimeter from the University of Colorado CU Sea Level Research Group. Individual tide-gauge records with enough length and quality can be used to compute the local sea level accelerating trend. However, the length should be longer than 60 years [4] to avoid large oscillations due to long-term tidal effects, which has also been confirmed by Parker himself [5]. As we known, individual tide-gauge records only represent the mean sea level variation within localized area. Parker presented the sea level records of several tide gauges in his Figure 1. The length of sea level time series are all longer than 100 years, and the 2nd order coefficients computed from the time series have both positive and negative values. The results show the largest positive acceleration in San Francisco, CA of +0.007 mm/a, and the largest negative acceleration in Honolulu, HI of 0.006 mm/a. These phenomena demonstrate that the local sea levels are accelerating or decelerating in different sea areas. Nevertheless, the GMSL accelerating trend is still unknown from his Figure 1. Either by tide gauges, or by satellite radar altimetry, many researches have achieved the conclusion that regional mean sea level trends are geographically dependent. Woodworth et al. [6] had pointed out that most of the tide-gauge data from Europe and North America display evidence for a positive acceleration, but the magnitude of the acceleration varies between stations. There are also a few of stations which display negative accelerations. From the most recent results published by AVISO (http://www.aviso.oceanobs.com/ en/news/ocean-indicators/mean-sea-level/) and that shown in Figure 1, we can see large difference, especially between eastern and western parts of the Pacific Ocean. Another fact is that the magnitude of the acceleration/deceleration depends on the time period of observations. For comparison, we select the tide gauges whose periods are longer then 60 years from the Revised Local Reference (RLR) dataset provided by PSMSL (http://www.psmsl.org/ data/obtaining/) [7], and remove the tide gauges whose spare time is greater then 30% to ensure the stability. Then considering the seasonal signals, the sea level records on each selected tide gauge are fitted by the following linear regression model: 2 cos(2 ) sin(2 ) cos(4 )

Mitigating errors in marine gravity recovery induced by sea level variability in SWOT observations