mithril, you are keeping me awake at night ;)The graph:From the...

  1. 13,639 Posts.
    lightbulb Created with Sketch. 39
    mithril, you are keeping me awake at night ;)

    The graph:
    From the wiki, it states that "This figure shows changes in sea level during the Holocene, the time following the end of the most recent glacial period, based on data from Fleming et al. 1998, Fleming 2000, & Milne et al. 2005. These papers collected data from various reports and adjusted them for subsequent vertical geologic motions, primarily those associated with post-glacial continental and hydroisostatic rebound."
    So it appears that it has already taken into account the GIA correction. Though what value was used I will have to research further

    The correction:
    "greenhart perhaps try some numbers, just note that GIA correction is an increase to the global mean sea level "rate". It in itself is like an acceleration(it would be 30mm/year/century in the wiki terms)"
    Suppose a car is accelerating in a straight line along a road, relative to us. Then one day someone discovers that the road itself is moving in the opposite direction to the car, relative to us. Is the car accelerating more or less? Neither, of course. Is the car actually moving faster than we originally measured? Yes. But this difference will be the same tomorrow, next year, and 100 years from now.

    Also, the 30 mm/year/centrury figure you mention is not an acceleration. It has the dimensions of distance. i.e. it is 30 mm, not 30 mm/year/century. And it is small compared with the values of sea level rise predicted in 50-100 years.

    "sea_level_rate(t+1) = f(sea_level_rate(t)) + acceleration(t)*(1 year) + correction."

    Ok, a nice start for a model. But the acceleration is very suspect. Some data indicate that it is zero or even negative. So let's leave the acceleration out altogether. But this is no reason to omit another term in the model that is unrelated to the acceleration.

    I see it this way:
    Models often have uncertaincies, and therefore best and worst cases.
    If someone repeatedly quotes the worst case, without also quoting the uncertainty, I am suspicious.
    If somone repeatedly quotes the best case, while emphasising the uncertainty, I am suspicious.

    The truth is probably somoewhere in between

    Meanwhile, we keep collecting data so that the model's uncertanties become smaller, or the model is abviously so flawed that it has to be revised.
 
arrow-down-2 Created with Sketch. arrow-down-2 Created with Sketch.