Bathythermograph Data Research Articles

New Insights into Tyrrhenian Sea Warming and Heat Penetration through Long-Term Expendable Bathythermograph Data

New Insights into Tyrrhenian Sea Warming and Heat Penetration through Long-Term Expendable Bathythermograph Data

Sensitivity of Observationally Based Estimates of Ocean Heat Content and Thermal Expansion to Vertical Interpolation Schemes

AbstractChanges in ocean heat content are a critical element of climate change, with the oceans containing about 90% of the excess heat stored in the climate system and 60% in the upper 700 db. Estimates of these changes are sensitive to horizontal mapping of the sparse historical data and errors in eXpendable BathyThermograph data. Here we show that they are also sensitive to the vertical interpolation of sparsely sampled data through the water column. We estimate, using carefully constructed vertical interpolation methods with high‐quality hydrographic (bottle and CTD) data, the observationally based upper ocean heat content increase (thermosteric sea level rise) from 1956 to 2020 is 285 Zeta Joules (0.55 mm yr−1), 14% (14%) larger than estimates relying on simple but biased linear interpolation schemes. The underestimates have a clear spatial pattern with their maximum near 15°N and 12°S, around the maxima in the curvature of the temperature‐depth profile.

Low-cost assimilation for sound speed fields in the PCA framework

Forecast ocean sound speed fields are often inaccurate and need to be reconciled with observation data. Conventional data-assimilation methods used for this are generally quite computationally intensive. A compressed representation of the forecast sound speed fields can be obtained using principal component analysis (PCA), where the forecast fields are represented by a linear combination fo PCA modes. We develop a low-cost assimilation approach that updates the PCA compressed representation of the background forecast field, based on observation data.The approach uses Bayes‘ rule to obtain the maximum likelihood estimate for the update in the space spanned by the PCA modes. Significant cost savings come from the dimensionality reduction that is provided by PCA. Results are presented for sound speed fields derived from HYCOM ocean forecast data and expendable bathythermograph data obtained from the Scipps Institution of Oceanography.

Quantifying Spread in Spatiotemporal Changes of Upper-Ocean Heat Content Estimates: An Internationally Coordinated Comparison

AbstractThe Earth system is accumulating energy due to human-induced activities. More than 90% of this energy has been stored in the ocean as heat since 1970, with ∼60% of that in the upper 700 m. Differences in upper-ocean heat content anomaly (OHCA) estimates, however, exist. Here, we use a dataset protocol for 1970–2008—with six instrumental bias adjustments applied to expendable bathythermograph (XBT) data, and mapped by six research groups—to evaluate the spatiotemporal spread in upper OHCA estimates arising from two choices: 1) those arising from instrumental bias adjustments and 2) those arising from mathematical (i.e., mapping) techniques to interpolate and extrapolate data in space and time. We also examined the effect of a common ocean mask, which reveals that exclusion of shallow seas can reduce global OHCA estimates up to 13%. Spread due to mapping method is largest in the Indian Ocean and in the eddy-rich and frontal regions of all basins. Spread due to XBT bias adjustment is largest in the Pacific Ocean within 30°N–30°S. In both mapping and XBT cases, spread is higher for 1990–2004. Statistically different trends among mapping methods are found not only in the poorly observed Southern Ocean but also in the well-observed northwest Atlantic. Our results cannot determine the best mapping or bias adjustment schemes, but they identify where important sensitivities exist, and thus where further understanding will help to refine OHCA estimates. These results highlight the need for further coordinated OHCA studies to evaluate the performance of existing mapping methods along with comprehensive assessment of uncertainty estimates.

Variations of the Kuroshio in the Luzon Strait Revealed by EOF Analysis of Repeated XBT Data and Sea‐Level Anomalies

AbstractThe Kuroshio in the Luzon Strait (KLS) has long been a focus of research for its complex ocean dynamics; however, it is poorly understood owing to a lack of long‐term in situ observations. Empirical orthogonal function (EOF) analyses were applied to quarterly expendable bathythermograph data (XBT, with a spatial resolution of 10–20 km in the Kuroshio region and 33 km in the other regions) and sea‐level anomaly (SLA) data, to examine the variations of the KLS. The first EOF modes of water temperature and SLA represent seasonal variations. The volume transport of the KLS referring to a 700 m depth reaches its maximum in warm months (19.4 Sv, July–September) and its minimum in cold months (17 Sv, January–March). This variation is similar to that east of Taiwan but is opposite to that east of Luzon, although the Luzon Strait is only 2° distant from the area east of Luzon. The outflow (1.3 Sv) from the South China Sea (SCS) to the Kuroshio in the warm months and the Kuroshio intrusion into the SCS in the cold months are responsible for this seasonal contrast. The second EOF modes of water temperature and SLA are related to the latitudinal movement of the Kuroshio recirculation on a time scale of 2–24 months. Lag correlation and wavelet analysis demonstrate that this mode is triggered by the westward‐propagating SLA along 21°–24°N and is affected by the downstream propagation of low‐frequency variation (mostly annual to biennial) from the North Equatorial Current bifurcation area.

Correcting Biases in Historical Bathythermograph Data Using Artificial Neural Networks

AbstractHistorical estimates of ocean heat content (OHC) are important for understanding the climate sensitivity of the Earth system and for tracking changes in Earth’s energy balance over time. Prior to 2004, these estimates rely primarily on temperature measurements from mechanical and expendable bathythermograph (BT) instruments that were deployed on large scales by naval vessels and ships of opportunity. These BT temperature measurements are subject to well-documented biases, but even the best calibration methods still exhibit residual biases when compared with high-quality temperature datasets. Here, we use a new approach to reduce biases in historical BT data after binning them to a regular grid such as would be used for estimating OHC. Our method consists of an ensemble of artificial neural networks that corrects biases with respect to depth, year, and water temperature in the top 10 m. A global correction and corrections optimized to specific BT probe types are presented for the top 1800 m. Our approach differs from most prior studies by accounting for multiple sources of error in a single correction instead of separating the bias into several independent components. These new global and probe-specific corrections perform on par with widely used calibration methods on a series of metrics that examine the residual temperature biases with respect to a high-quality reference dataset. However, distinct patterns emerge across these various calibration methods when they are extrapolated to BT data that are not included in our cross-instrument comparison, contributing to uncertainty that will ultimately impact estimates of OHC.

The Observed Seasonal Cycle of Macronutrients in Drake Passage: Relationship to Fronts and Utility as a Model Metric

AbstractThe Drake Passage Time‐series (DPT) is used to quantify the spatial and seasonal variability of historically undersampled, biogeochemically relevant properties across the Drake Passage. From 2004–2017, discrete ship‐based observations of surface macronutrients (silicate, nitrate, and phosphate), temperature, and salinity have been collected 5–8 times per year as part of the DPT program. Using the DPT and Antarctic Circumpolar Current (ACC) front locations derived from concurrent expendable bathythermograph data, the distinct physical and biogeochemical characteristics of ACC frontal zones are characterized. Biogeochemical‐Argo floats in the region confirm that the near‐surface sampling scheme of the DPT robustly captures mixed‐layer biogeochemistry. While macronutrient concentrations consistently increase toward the Antarctic continent, their meridional distribution, variability, and biogeochemical gradients are unique across physical ACC fronts, suggesting a combination of physical and biological processes controlling nutrient availability and nutrient front location. The Polar Front is associated with the northern expression of the Silicate Front, marking the biogeographically relevant location between silicate‐poor and silicate‐rich waters. South of the northern Silicate Front, the silicate‐to‐nitrate ratio increases, with the sharpest gradient in silicate associated with the Southern ACC Front (i.e., the southern expression of the Silicate Front). Nutrient cycling is an important control on variability in the surface ocean partial pressure of carbon dioxide (pCO2). The robust characterization of the spatiotemporal variability of nutrients presented here highlights the utility of biogeochemical time series for diagnosing and potentially reducing biases in modeling Southern Ocean pCO2 variability, and by inference, air‐sea CO2 flux.

Geostatistical Inversion of Seismic Oceanography Data for Ocean Salinity and Temperature Models

Conventional multi-channel seismic reflection data, known as seismic oceanography, has recently been used for the qualitative interpretation of meso- to large-scale hydrographic structures of interest. Seismic oceanography has been successfully imaging oceanographic structures in an intermediate scale not sampled by traditional oceanographic tools, such as conductivity, depth and temperature measurements and eXpendable BathyThermograph (XBT) data. However, few attempts have been made for successfully quantifying ocean properties, such as ocean temperature and salinity, directly from the seismic reflection data. This work presents an iterative geostatistical methodology capable of inverting conventional seismic oceanographic data simultaneously for high-resolution temperature and salinity ocean models. The proposed methodology was developed and implemented in a real set of contemporaneous XBT data and two-dimensional seismic profile acquired southwest of Portugal. The resulting high-resolution temperature and salinity models reproduce existing XBT data not used to constrain the geostatistical inversion, which permits reliable quantification of the ocean properties of interest.

The Impact of Improved Thermistor Calibration on the Expendable Bathythermograph Profile Data

AbstractExpendable bathythermograph (XBT) data provide one of the longest available records of upper-ocean temperature. However, temperature and depth biases in XBT data adversely affect estimates of long-term trends of ocean heat content and, to a lesser extent, estimates of volume and heat transport in the ocean. Several corrections have been proposed to overcome historical biases in XBT data, which rely on constantly monitoring these biases. This paper provides an analysis of data collected during three recent hydrographic cruises that utilized different types of probes, and examines methods to reduce temperature and depth biases by improving the thermistor calibration and reducing the mass variability of the XBT probes.The results obtained show that the use of individual thermistor calibration in XBT probes is the most effective calibration to decrease the thermal bias, improving the mean thermal bias to less than 0.02°C and its tolerance from 0.1° to 0.03°C. The temperature variance of probes with screened thermistors is significantly reduced by approximately 60% in comparison to standard probes. On the other hand, probes with a tighter weight tolerance did not show statistically significant reductions in the spread of depth biases, possibly because of the small sample size or the sensitivity of the depth accuracy to other causes affecting the analysis.

Variability of the subtropical mode water in the Southwest Pacific

AbstractThe variability of Subtropical Mode Water (STMW) in the Southwest Pacific is investigated using a 28 year‐long time series (1986–2014) of high‐resolution expendable bathythermograph data north of New Zealand (PX06) and a shorter time series, the Roemmich‐Gilson monthly Argo optimal interpolation for the 2004–2014 period. The variability in STMW inventories is compared to the variability in air‐sea heat fluxes, mixed layer depths and transport of the East Auckland Current (EAUC) to assess both the atmospheric and oceanic roles influencing the formation and decay of STMW. The STMW north of New Zealand has a short lifespan with little persistence of the water mass from 1 year to the next one. Deeper mixed layers and negative anomalies in surface heat fluxes are correlated with increased formation of STMW. The heat content of the STMW layer is anticorrelated with inventories, particularly during the El Niño years. This suggests that large volumes of STMW are coincident with cooler conditions in the prior winter and less oceanic heat storage. There is significant seasonal and interannual variability in STMW inventories, however there are no trends in STMW properties, including its core layer temperature over the last decade. The variability of the winter EAUC transport is highly correlated with the STMW inventories and thermocline depth in the following spring, suggesting ocean dynamics deepen the thermocline and precondition for deeper mixed layers.

XBT Science: Assessment of Instrumental Biases and Errors

Abstract Expendable bathythermograph (XBT) data were the major component of the ocean temperature profile observations from the late 1960s through the early 2000s, and XBTs still continue to provide critical data to monitor surface and subsurface currents, meridional heat transport, and ocean heat content. Systematic errors have been identified in the XBT data, some of which originate from computing the depth in the profile using a theoretically and experimentally derived fall-rate equation (FRE). After in-depth studies of these biases and discussions held in several workshops dedicated to discussing XBT biases, the XBT science community met at the Fourth XBT Science Workshop and concluded that XBT biases consist of 1) errors in depth values due to the inadequacy of the probe motion description done by standard FRE and 2) independent pure temperature biases. The depth error and temperature bias are temperature dependent and may depend on the data acquisition and recording system. In addition, the depth bias also includes an offset term. Some biases affecting the XBT-derived temperature profiles vary with manufacturer/probe type and have been shown to be time dependent. Best practices for historical XBT data corrections, recommendations for future collection of metadata to accompany XBT data, impact of XBT biases on scientific applications, and challenges encountered are presented in this manuscript. Analysis of XBT data shows that, despite the existence of these biases, historical XBT data without bias corrections are still suitable for many scientific applications, and that bias-corrected data can be used for climate research.

Mapping the Antarctic Polar Front: weekly realizations from 2002 to 2014

Abstract. We map the weekly position of the Antarctic Polar Front (PF) in the Southern Ocean over a 12-year period (2002–2014) using satellite sea surface temperature (SST) estimated from cloud-penetrating microwave radiometers. Our study advances previous efforts to map the PF using hydrographic and satellite data and provides a unique realization of the PF at weekly resolution across all longitudes (doi:10.1594/PANGAEA.855640). The mean path of the PF is asymmetric; its latitudinal position spans from 44 to 64° S along its circumpolar path. SST at the PF ranges from 0.6 to 6.9 °C, reflecting the large spread in latitudinal position. The average intensity of the front is 1.7 °C per 100 km, with intensity ranging from 1.4 to 2.3 °C per 100 km. Front intensity is significantly correlated with the depth of bottom topography, suggesting that the front intensifies over shallow bathymetry. Realizations of the PF are consistent with the corresponding surface expressions of the PF estimated using expendable bathythermograph data in the Drake Passage and Australian and African sectors. The climatological mean position of the PF is similar, though not identical, to previously published estimates. As the PF is a key indicator of physical circulation, surface nutrient concentration, and biogeography in the Southern Ocean, future studies of physical and biogeochemical oceanography in this region will benefit from the provided data set.

Aircraft Observations of Dry Air, the ITCZ, Convective Cloud Systems, and Cold Pools in MJO during DYNAMO

Abstract One of the most challenging problems in predicting the Madden–Julian oscillation (MJO) is the initiation of large-scale convective activity associated with the MJO over the tropical Indian Ocean. The lack of observations is a major obstacle. The Dynamics of the MJO (DYNAMO) field campaign collected unprecedented observations from air-, land-, and ship-based platforms from October 2011 to February 2012. Here we provide an overview of the aircraft observations in DYNAMO, which captured an MJO initiation event from November to December 2011. The National Oceanic and Atmospheric Administration (NOAA) WP-3D aircraft was stationed at Diego Garcia and the French Falcon 20 aircraft on Gan Island in the Maldives. Observations from the two aircraft provide a unique dataset of three-dimensional structure of convective cloud systems and their environment from the flight level, airborne Doppler radar, microphysics probes, ocean surface imaging, global positioning system (GPS) dropsonde, and airborne expendable bathythermograph (AXBT) data. The aircraft observations revealed interactions among dry air, the intertropical convergence zone (ITCZ), convective cloud systems, and air–sea interaction induced by convective cold pools, which may play important roles in the multiscale processes of MJO initiation. This overview focuses on some key aspects of the aircraft observations that contribute directly to better understanding of the interactions among convective cloud systems, environmental moisture, and the upper ocean during the MJO initiation over the tropical Indian Ocean. Special emphasis is on the distinct characteristics of convective cloud systems, environmental moisture and winds, air–sea fluxes, and convective cold pools during the convectively suppressed, transition/onset, and active phases of the MJO.

Interannual variability of the Indonesian Throughflow transport: A revisit based on 30 year expendable bathythermograph data

AbstractBased on 30 year repeated expendable bathythermograph (XBT) deployments between Fremantle, Western Australia, and the Sunda Strait, Indonesia, from 1984 to 2013, interannual variability of geostrophic transport of the Indonesian Throughflow (ITF) and its relationships with El Niño Southern Oscillation (ENSO) and the Indian Ocean Dipole (IOD) are investigated. The IOD induced coastal Kelvin waves propagate along the Sumatra‐Java coast of Indonesia, and ENSO induced coastal Kelvin waves propagate along the northwest coast of Australia, both influencing interannual variations of the ITF transport. The ITF geostrophic transport is stronger during La Niña phase and weaker during El Niño phase, with the Niño3.4 index leading the ITF variability by 7 months. The Indian Ocean wind variability associated with the IOD to a certain extent offset the Pacific ENSO influences on the ITF geostrophic transport during the developing and mature phases of El Niño and La Niña, due to the covarying IOD variability with ENSO. The ITF geostrophic transport experiences a strengthening trend of about 1 Sv every 10 years over the study period, which is mostly due to a response to the strengthening of the trade winds in the Pacific during the climate change hiatus period. Decadal variations of the temperature‐salinity relationships need to be considered when estimating the geostrophic transport of the ITF using XBT data.

The impact of historical biases on the XBT‐derived meridional overturning circulation estimates at 34°S

AbstractAn observational system based on high‐density expendable bathythermograph (XBT) data has provided the longest record of the South Atlantic meridional overturning and heat transport estimates across 34°S. Measurement biases are a point of concern for the capability of an XBT system to capture long‐term trends in volume and heat transports, and the impact of such biases on the meridional overturning estimates has never been quantified. In the present study, the sensitivity of the meridional overturning circulation and heat transport to uncertainties in XBT measurements is quantified under the framework of an eddy‐resolving model simulation. Results show that XBT measurement biases after 2010 can translate into small meridional overturning errors on the order of 3% or 0.38 Sv (1Sv = 106 m3 s−1), and 0.025 PW (1 PW = 1015 W) or 8% of the meridional heat transport in the model. Historical XBT‐derived trends in transport estimates across 34°S are stronger and statistically significant after the late 1990s, 0.3 Sv decade−1 and 0.02 PW decade−1. These trends are mostly due to the XBT linear depth bias, with smaller contributions associated with temperature and depth offsets from the historical record. Long‐term trends calculated from Simple Ocean Data Assimilation reanalysis, estimated as 0.1 Sv/decade and 0.006 PW/decade, are 3 times smaller than the XBT‐derived historical trends. Therefore, an adequate correction of historical XBT data is necessary for an early detection of trends in the meridional overturning circulation and heat transport.

Time, Probe Type, and Temperature Variable Bias Corrections to Historical Expendable Bathythermograph Observations

Abstract Systematic biases in historical expendable bathythermograph (XBT) data are examined using two datasets: 4151 XBT–CTD side-by-side pairs from 1967 to 2011 and 218 653 global-scale XBT–CTD pairs (within one month and 1°) extracted from the World Ocean Database 2009 (WOD09) from 1966 to 2010. Using the side-by-side dataset, it was found that both the pure thermal bias and the XBT fall rate (from which the depth of observation is calculated) increase with water temperature. Correlations between the terminal velocity A and deceleration B terms of the fall-rate equation (FRE) and between A and the offset from the surface terms are obtained, with A as the dominant term in XBT fall-rate behavior. To quantify the time variation of the XBT fall-rate and pure temperature biases, global-scale XBT–CTD pairs are used. Based on the results from the two datasets, a new correction scheme for historical XBT data is proposed for nine independent probe-type groups. The scheme includes corrections for both temperature and depth records, which are all variable with calendar year, water temperature, and probe type. The results confirm those found in previous studies: a slowing in fall rate during the 1970s and 2000s and the large pure thermal biases during 1970–85. The performance of nine different correction schemes is compared. After the proposed corrections are applied to the XBT data in the WOD09 dataset, global ocean heat content from 1967 to 2010 is reestimated.

Biases in Expendable Bathythermograph Data: A New View Based on Historical Side-by-Side Comparisons

Abstract Because they make up 56% of ocean temperature profile data between 1967 and 2001, quantifying the biases in expendable bathythermograph (XBT) data is fundamental to understanding the evolution of the planetary energy and sea level budgets over recent decades. The nature and time history of these biases remain in dispute and dominate differences in analyses of the history of ocean warming. A database of over 4100 side-by-side deployments of XBTs and conductivity–temperature–depth (CTD) data has been assembled, and this unique resource is used to characterize and separate out the pure temperature bias from depth error in a way that was not previously possible. Two independent methods of bias extraction confirm that the results are robust to bias model and fitting method. It was found that there is a pure temperature bias in Sippican probes of ~0.05°C, independent of depth. The temperature bias has a time dependency, being larger (~0.1°C) in the earlier analog acquisition era and being likely due to changes in recorder type. Large depth errors are found in the 1970s–80s in shallower-measuring Sippican T4/T6 probe types, but the deeper-measuring Sippican T7/Deep Blue (DB) types have no error during this time. The Sippican T7/DB fall rate slows from ~1990 onward. It is found that year-to-year variations in fall rate have a bigger effect on corrections to the global XBT database than do any small effects of ocean temperature on fall rate. This study has large implications for the future development of better schemes to correct the global historical XBT archive.

South Atlantic meridional fluxes

The properties of the meridional overturning circulation (MOC) and associated meridional heat transport (MHT) and salt fluxes are analyzed in the South Atlantic. The oceanographic data used for the study consist of Expendable bathythermograph (XBT) data collected along 27 sections at nominally 35°S for the period of time 2002–2011, and Argo profile data collected in the region. Previous estimates obtained with a shorter record are improved and extended, using new oceanographic sections and wind fields. Different wind products are analyzed to determine the uncertainty in the Ekman component of the MHT derived from their use. Results of the analysis provide a 9-year time series of MHT, and volume transport in the upper layer of the MOC. Salt fluxes at 35°S are estimated using a parameter introduced by numerical studies, the Mov that represents the salt flux and helps determine the basin scale salt feedback associated with the MOC. Volume and heat transport by the western and eastern boundary currents are estimated, and their covariablity is examined. Analysis of the data shows that the South Atlantic is responsible for a northward MHT with a mean value of 0.54±0.14PW. The MHT exhibits no significant trend from 2002 to 2011. The MOC varies from 14.4 to 22.7Sv with a mean value of 18.1±2.3Sv and the maximum overturning transport is found at a mean depth of 1250m. Statistical analysis suggests that an increase of 1Sv in the MOC leads to an increase of the MHT of 0.04±0.02PW. Estimates of the Mov from data collected from three different kinds of observations, contrary to those obtained from models, feature a positive salt advection feedback (Mov<0) suggesting that freshwater perturbations will be amplified and that the MOC is bistable. In other words, the MOC might collapse with a large enough freshwater perturbation. Observations indicate that the mean value of the Brazil Current is −8.6±4.1Sv at 24°S and −19.4±4.3Sv at 35°S, increasing towards the south. East of 3°E, the northward flowing Benguela Current and Agulhas rings have a net northward transport of 22.5±4.7Sv. No significant correlation is observed between the MOC and the Brazil Current transport, and most of the compensation derives from the eastern boundary and interior transports. Products from the Ocean general circulation model For the Earth Simulator (OFES) are used to validate methodology used to extend the XBT record, and to aid in the interpretation of the observed findings.

World ocean heat content and thermosteric sea level change (0–2000 m), 1955–2010

We provide updated estimates of the change of ocean heat content and the thermosteric component of sea level change of the 0–700 and 0–2000 m layers of the World Ocean for 1955–2010. Our estimates are based on historical data not previously available, additional modern data, and bathythermograph data corrected for instrumental biases. We have also used Argo data corrected by the Argo DAC if available and used uncorrected Argo data if no corrections were available at the time we downloaded the Argo data. The heat content of the World Ocean for the 0–2000 m layer increased by 24.0 ± 1.9 × 1022 J (±2S.E.) corresponding to a rate of 0.39 W m−2 (per unit area of the World Ocean) and a volume mean warming of 0.09°C. This warming corresponds to a rate of 0.27 W m−2 per unit area of earth's surface. The heat content of the World Ocean for the 0–700 m layer increased by 16.7 ± 1.6 × 1022 J corresponding to a rate of 0.27 W m−2(per unit area of the World Ocean) and a volume mean warming of 0.18°C. The World Ocean accounts for approximately 93% of the warming of the earth system that has occurred since 1955. The 700–2000 m ocean layer accounted for approximately one‐third of the warming of the 0–2000 m layer of the World Ocean. The thermosteric component of sea level trend was 0.54 ± .05 mm yr−1 for the 0–2000 m layer and 0.41 ± .04 mm yr−1 for the 0–700 m layer of the World Ocean for 1955–2010.

The effect of interpolation methods in temperature and salinity trends in the Western Mediterranean

Temperature and salinity data in the historical record are scarce and unevenly distributed in space and time and the estimation of linear trends is sensitive to different factors. In the case of the Western Mediterranean, previous works have studied the sensitivity of these trends to the use of bathythermograph data, the averaging methods or the way in which gaps in time series are dealt with. In this work, a new factor is analysed: the effect of data interpolation. Temperature and salinity time series are generated averaging existing data over certain geographical areas and also by means of interpolation. Linear trends from both types of time series are compared. There are some differences between both estimations for some layers and geographical areas, while in other cases the results are consistent. Those results which do not depend on the use of interpolated or non-interpolated data, neither are influenced by data analysis methods can be considered as robust ones. Those results influenced by the interpolation process or the factors analysed in previous sensitivity tests are not considered as robust results.