Against my better judgement, I'm going to attempt to educate you. To preface, I have a BS in Mechanical Engineering, a PhD in Genetics, and have experience in real world mathematical modeling, big data analytics, and advanced statistical analyses. My first publication in graduate school is the highest cited paper in my Columbia-graduated advisor's entire career.
1. Climate scientists aren't scientists. They are wannabe mathematicians that are bad at math, as climate science applicants have appallingly low average quantitative GRE scores that are inadequate for admission to hard science programs. Calling them scientists is an insult to REAL scientists.
2. The climate is a massively multivariate system akin in scale and complexity to the human body. Saying that we can accurately model the climate is like saying we can accurately model exactly how a person will age, quantitatively predict disease manifestation, or pre-determine the effect of a new drug or genetic change on the entire body. We have clinical trials for a reason.
3. The models used by climate science are derived from a number of different data sets, with vast differences in error variance. Combining the sets together creates a statistical nightmare with error tolerances so large that the model is essentially useless.
a. Satellite data is generally the most accurate, and is limited to less than 100 years of data.
b. Thermometer data is typically derived from airports, which has to be normalized on a time gradient because of the expansion of big cities and the heat island effect. This is rarely done in studies. In addition, there isn't a central starting point in time for creating a baseline global average, as airports were fewer in number early on and largely concentrated in North America and Europe. As a result, this data prior to maybe the 1960's (and possibly later) is going to have to either be extrapolated to other parts of the world based on other data sources (i.e. ice cores), or else missing crucial data points. Either way, a very significant amount of error tolerance is present in the set.
c. Water based temp data. There are 2 types - ship based ocean surface readings, and buoy based ocean subsurface readings. Surface data is widely accepted as unreliable due to a roughly 2 degree artificially inflated reading due to light reflection off the surface. Notoriously, buoy based data was removed from the NOAA report under Obama that was presented to the UN, as the initial results including it didn't fit the narrative.
d. Ice Cores. Horribly inaccurate with huge error variance. This variance also scales exponentially the further you go back in time. In addition, they are limited to areas near the poles, so they are a very poor indicator of historical average global temperatures. As they are the only source of somewhat historical data, they constitute a very very poor "control" to base a climate model on.
e. CO2 data. This comes from either weather balloons, which started somewhere in the mid last century, or ice cores which as noted above are horribly unreliable as a data set. As far as I'm aware, there hasn't even been a control study done to assess how accurate trapped CO2 readings are over time. That would require measuring CO2 in atmosphere at site of ice cores, and then taking an ice core sample every year to measure what is trapped in ice. You would need at least a 50 year study to even begin to validate ice cores as a reliable source for CO2 data, and really not even then since layers become compacted together when you are looking at anything beyond 100 years or so. Without that control study, there is no way to normalize the data to account for less than 100% CO2 capture. As a result, it will artificially look like CO2 is rising over time.
f. Size of Data Set. The Earth is hundreds of millions of years old, and we only have data that goes back at best about 10,000 years with any degree of usefulness. There are a few additional data points derived from geological records and older ice core data tied to known historical events (mass extinctions, eruption of Pompeii, etc), but there is no way to validate the accuracy of the readings and they are too far in between to build any sort of ancient data set. This is especially problematic because the Earth goes through sustained periods of heating and cooling as the axis wobbles. It is widely believed that we are actually near the peak in a warming cycle, but this is not properly reflected in the climate models. In addition, the Earth's magnetic poles shift on occasion, with far far more impact on the climate than man could ever dream of causing.
4. Even if all the data was 100% accurate, you still have to show that the climate change is man made.
a. Their only "proof" of this is to correlate temperature change to CO2/greenhouse gas level change. First, that correlation doesn't equal causation is a cornerstone of scientific research. Second, that correlation is heavily dependent on how the data is represented and sourced (ex: how average global Temps are calculated). The correlation could be supported by a series of site specific t-tests and correlations subsequently analyzed together in both One Way ANOVA and MANOVA (i.e. showing a statistically significant correlation independently at a variety of cities/locations), but to my knowledge this hasn't been done.
b. How do you accurately attribute how much is man made? You would need to be able to exactly calculate the amount of greenhouse gas produced by man vs nature, and there is simply no way currently to do this. Everything is based on a flawed aggregate of calculated estimates rife with more error variance (plant emissions, car emissions, cow farts, etc). In addition, these models have either done a poor job of modeling or else left out entirely the effect of volcanoes, solar flares, and sub-sea magma vents, which is a MASSIVE source of both heat and CO2.
5. Climate science has been so heavily politicized that the scientific method and ethics have been abandoned. There have been several examples of data tampering (most notably by the NOAA), and any academic that tries to publish a study that runs contrary to the established narrative is ostracized. The formerly well respected researcher at Georgia Tech comes to mind, in particular.
6. In conclusion, there simply isn't enough data to to build an accurate model of our climate or determine how much change is attributed to man. There is too much variance and too high of error tolerances to achieve statistically significant results. The models are built by highly politicized, unethical academics that aren't even good at math. I would be absolutely ashamed to attribute my name to any of these pseudoscience "studies".
