Of Models And Melting Ice Caps

Once again climate scientists have put forth a scary prediction about melting ice caps and once again they are portraying output from computer models as a reliable prediction of things to come. The normally staid and reliable journal Nature emblazoned its cover with “Rising tide” in large black print, with the subtext “A 500-year model of Antarctica's contribution to future sea-level rise” in smaller print below. When a closer look is taken at this new model it turns out to be a house of cards, incomplete and built on top of other climate change models that are known to be faulty. As the old saying goes: garbage in, garbage out. Yet a major science journal chose this piece of computerized legerdemain as its lead article. No wonder that climate alarmists, and climate science in general, have fallen into such ill repute.

In the news section of the March 31, 2016, issue of Nature, and article proclaims “Antarctic model raises prospect of unstoppable ice collapse.” Penned by science reporter Jeff Tollefson, it breathlessly spreads the word about impending, catastrophic collapse of the Antarctic ice-sheets:

Choices that the world makes this century could determine the fate of the massive Antarctic ice sheet. A study published online this week in Nature finds that continued growth in greenhouse-gas emissions over the next several decades could trigger an unstoppable collapse of Antarctica’s ice — raising sea levels by more than a metre by 2100 and more than 15 metres by 2500.

The article continues by quoting one of the authors of the associated paper, co-author Rob DeConto, a geoscientist at the University of Massachusetts Amherst. He says this regarding the impact of such massive sea-level change: “That is literally remapping how the planet looks from space.” The good news, says DeConto, is that it projects little or no sea-level rise from Antarctic melt if greenhouse-gas emissions are reduced quickly enough to limit the average global temperature rise to about 2°C. That last bit reveals all. This is yet another article justifying the arbitrary “limit” agreed upon by politicians and UN bureaucrats at the last in a series of climate boondoggles.

In “Contribution of Antarctica to past and future sea-level rise,” authors DeConto and Pollard report on potential changes in global mean sea-level (GMSL) based on Antarctic ice-sheet simulations driven by a time-evolving, proxy-based atmosphere–ocean climatology. In other words, a model. Here is the abstract:

Polar temperatures over the last several million years have, at times, been slightly warmer than today, yet global mean sea level has been 6–9 metres higher as recently as the Last Interglacial (130,000 to 115,000 years ago) and possibly higher during the Pliocene epoch (about three million years ago). In both cases the Antarctic ice sheet has been implicated as the primary contributor, hinting at its future vulnerability. Here we use a model coupling ice sheet and climate dynamics—including previously underappreciated processes linking atmospheric warming with hydrofracturing of buttressing ice shelves and structural collapse of marine-terminating ice cliffs—that is calibrated against Pliocene and Last Interglacial sea-level estimates and applied to future greenhouse gas emission scenarios. Antarctica has the potential to contribute more than a metre of sea-level rise by 2100 and more than 15 metres by 2500, if emissions continue unabated. In this case atmospheric warming will soon become the dominant driver of ice loss, but prolonged ocean warming will delay its recovery for thousands of years.

Note that they say that atmospheric warming is the culprit here, we will return to that later. Together, DeConto and Pollard conceived the model experiments, developed the models, and wrote up their results in this paper. They calibrated their model against estimates for sea-level changes in the past, using estimates of both the sea-level and atmospheric CO2 concentrations from proxy records. Little mention is made of the uncertainty inherent in such measurements. Using data from the Pliocene and Last Interglacial (LIG, between 130,000 and 115,000 years ago) the researchers calibrated their model so that it reproduced their best guess at what the historical conditions and responses were. But the important results here are projections for the future. These are shown in the figure below.

Using the same model physics and parameter values as used in the Pliocene and LIG simulations, the ice-sheet model was applied to long-term future simulations. Atmospheric forcing—in other words, air temperature—was provided by high-resolution RCM simulations following three extended Representative Carbon Pathway (RCP) scenarios (RCP2.6, RCP4.5 and RCP8.5)36. As the authors state: “Future circum-Antarctic ocean temperatures used in our time-evolving sub-ice melt-rate calculations come from matching, high-resolution (1°) National Center for Atmospheric Research (NCAR) CCSM4 simulations. The simulations begin in 1950 to provide some hindcast spinup, and are run for 550 years to 2500.”

The problem here is that the models being used to provide environmental temperature projections for 500 years into the future have been shown to not be very good at their jobs. In fact, they have been shown to run consistently hot, predicting temperatures far above that which actually occur. The only reason they appear to have any accuracy at all is that the models are continually tweaked to match past data, that's the hindcasting mentioned above. Some times the temperature data are “adjusted” as well. No amount of this constant readjustment has managed to make such models accurate.

The authors even acknowledge that one of the input generating models has a cold bias. “The CCSM4 simulations providing the model’s sub-ice-shelf melt rates underestimate the penetration of warm Circum-Antarctic Deep Water into the Amundsen and Bellingshausen seas observed in recent decades,” they state. “As a result, the model fails to capture recent, 21st-century thinning and grounding-line retreat along the southern Antarctic Peninsula and the Amundsen Sea Embayment.” Interesting that they admit no other bias in the models used to drive their ice melt model.

And therein lies the problem with reporting model output as though it is real science. Even scientists directly involved with climate modeling admit that models are not accurate representations of Earth's climate system. Recently, Eduardo Zorita, a paleoclimatologist at the GKSS Research Centre in Germany, told Spiegel that climate models have a long way to go. According to this IPCC climate modeler, much remains poorly understood and today's models are “hardly trustworthy.” And one of the things that the models are worst at is changes in precipitation.

Since the ice-sheet model results were published another paper, also in Nature, described a reconstruction of 1,200 years of water's history in the Northern Hemisphere, based on proxy data. According to “Water's past revisited to predict its future ,” the paper by Fredrik C. Ljungqvist and colleagues, “fuels the debate about whether anthropogenic climate change affected twentieth-century precipitation.”

In “Northern Hemisphere hydroclimate variability over the past twelve centuries,” Ljungqvist et al. took a long hard look at proxy data and coupled atmosphere–ocean general circulation model simulations. Their conclusions are at the end of this rather wordy abstract describing their study:

Accurate modelling and prediction of the local to continental-scale hydroclimate response to global warming is essential given the strong impact of hydroclimate on ecosystem functioning, crop yields, water resources, and economic security. However, uncertainty in hydroclimate projections remains large, in part due to the short length of instrumental measurements available with which to assess climate models. Here we present a spatial reconstruction of hydroclimate variability over the past twelve centuries across the Northern Hemisphere derived from a network of 196 at least millennium-long proxy records. We use this reconstruction to place recent hydrological changes and future precipitation scenarios in a long-term context of spatially resolved and temporally persistent hydroclimate patterns. We find a larger percentage of land area with relatively wetter conditions in the ninth to eleventh and the twentieth centuries, whereas drier conditions are more widespread between the twelfth and nineteenth centuries. Our reconstruction reveals that prominent seesaw patterns of alternating moisture regimes observed in instrumental data across the Mediterranean, western USA, and China have operated consistently over the past twelve centuries. Using an updated compilation of 128 temperature proxy records, we assess the relationship between the reconstructed centennial-scale Northern Hemisphere hydroclimate and temperature variability. Even though dry and wet conditions occurred over extensive areas under both warm and cold climate regimes, a statistically significant co-variability of hydroclimate and temperature is evident for particular regions. We compare the reconstructed hydroclimate anomalies with coupled atmosphere–ocean general circulation model simulations and find reasonable agreement during pre-industrial times. However, the intensification of the twentieth-century-mean hydroclimate anomalies in the simulations, as compared to previous centuries, is not supported by our new multi-proxy reconstruction. This finding suggests that much work remains before we can model hydroclimate variability accurately, and highlights the importance of using palaeoclimate data to place recent and predicted hydroclimate changes in a millennium-long context.

In the text of the paper the unreliability of current climate models is expanded on. They know the models are wrong, and they don't even know why they are wrong.

There are several possible explanations for the limited ability of climate models to simulate hydroclimate variability accurately: knowledge gaps in describing physical processes defining model sub-grid parameterizations, systematic absolute temperature biases in the models affecting evaporation and precipitation processes, and an underestimation of internal hydroclimate variability.

This, in turn, is important because yet another paper, “Glacial fluctuations since the ‘Medieval Warm Period’ at Rothera Point (western Antarctic Peninsula),” published in The Holocene, has found that significant glacial advance and retreat occurs due to fluctuation in precipitation. “[D]espite the recent air warming rate being higher, the glacial retreat rate during the MWP was similar to the present, suggesting that increased snow accumulation in recent decades may have counterbalanced the higher warming rate.”

What this means is that snow accumulation has a large impact on glacial advance and retreat. It is not just the rate of melting that counts, you also have to account for precipitation changes which can lead to accelerated ice accumulation. More simply put, it doesn't matter how fast an ice-sheet is melting if it is growing even faster. Like that old Moody Blues album, it's a question of balance. This directly challenges the 500-year model study's conclusion that warming air dominates ice melt. It's not warm air, it's precipitation that's the dominant factor. The further implication of all this is that a new computer model, which invariably does not contain all of the factors involved in ice-sheet melting, being feed by other models what are hardly reliable or accurate, cannot be expected to accurately predict the future.

Scientists actually doing real science have taken the astounding step of measuring the ice balance in Antarctica. A 2015 NASA study says that an increase in Antarctic snow accumulation that began 10,000 years ago is currently adding enough ice to the continent to outweigh the increased losses from its thinning glaciers. According to analysis of satellite data, the Antarctic ice sheet showed a net gain of 112 billion tons of ice a year from 1992 to 2001. That net gain slowed to 82 billion tons of ice per year between 2003 and 2008. So much for the conclusions of other studies, including the Intergovernmental Panel on Climate Change’s (IPCC) 2013 report, which says that Antarctica is overall losing land ice.

NASA is not alone in documenting ice gain in Antarctica. according to a study published in the journal Geophysical Research Letters. As reported in a press release from the British Antarctic Survey, annual snow accumulation on West Antarctica’s coastal ice sheet increased dramatically during the 20th century. The new study used ice cores to estimate annual snow accumulation from 1712 to 2010 along the coastal West Antarctic. Until 1899, annual snow accumulation remained steady, averaging 33 and 40 centimeters (13 and 16 inches) water, or melted snow, each year at two locations. Annual snow accumulation increased in the early 20th century, rising 30% between 1900 and 2010. More tellingly, the researchers found that in the last 30 years, the ice sheet gained nearly 5 meters (16 feet) more water than it did during the first 30 years of the time period examined. So ice sheet growth has been measured by two different methods: satellite and ice core measurements.

What was the prevailing view before the alarmist 500-year model was reported? A different Nature study, published in December, 2015, suggested that Antarctic melting was unlikely to produce more than 30 centimeters of sea-level rise by 2100. Significantly less than the 1.6 meters predicted by the new model. While noting that newly identified processes such as surface melting and the collapse of ice cliffs could increase ice loss, the older paper by Catherine Ritz et al., “Potential sea-level rise from Antarctic ice-sheet instability constrained by observations,” was built on observation. That is, actual science, not computerized hand-waving.

DeConto and Pollard said they used “highly resolved atmosphere and ocean model components rather than intermediate-complexity climate models or simplified climate forcing, and calibration based on major retreat during warm palaeoclimates rather than recent minor retreat driven by localized ocean forcing.” Except that those models are still inaccurate, their calibration has been done using proxy measurements with sizable intrinsic uncertainty, and the important contribution of precipitation and surface accumulation is not, can not, be addressed by current models.

To do science you need you need three things: an hypothesis that explains some natural phenomenon, a set of predictions that allow the hypothesis to be tested, and empirical measurements that either validate or disprove the hypothesis. Models merely automate the prediction making phase, they are not proof of anything. The potential catastrophic results cited by the authors require a huge collapse of the Antarctic ice-sheet, something never observed and hence totally speculative. Unless Nature has changed its name to WAG (Wild Ass Guess) this should not have been published let alone put on the cover. In short, this new result may be interesting modeling but it is not science.

Be safe, enjoy the interglacial and stay skeptical.

Climate Modeling Dominates Climate Science

In a stunning new paper, "Climate Modeling Dominates Climate Science," by Patrick J. Michaels and David E. Wojick, the extent of over reliance on climate modeling in climate science has been exposed. The research paper surveyed the entire literature of science for the last ten years, using Google Scholar, looking for modeling. They found that climate change science accounts for fully 55% of the modeling done in all of science. Quoting the article:

In fact the number of climate change articles that include one of the three modeling terms is 97% of those that just include climate change. This is further evidence that modeling completely dominates climate change research.

This shows how fake climate science "research" really is while at the same time tarnishing the reputation of computer modeling, which is a useful tool when applied properly.