The evolution of scaling laws in the sea ice floe size distribution
The evolution of scaling laws in the sea ice floe size distribution
复制标题
DOI：
10.1002/2016jc012573
复制
发表时间：
2017-09-01
影响因子：
3.6
通讯作者：
Tziperman, Eli
中科院分区：
文献类型：
--
作者：
Horvat, Christopher;Tziperman, Eli
亚网格尺度浮冰的大小和厚度分布（FSTD）是一个新兴的气候变量，在海冰、海洋和大气之间的耦合中起着主导作用。FSTD，然而，是很难测量的水平尺度的个别浮冰，导致普遍使用的幂律缩放来描述it. Evolutionary耦合的混合层FSTD模型的一个典型的边缘冰区明确模拟在这里，开发一个更深入的了解如何在浮冰规模的过程中活跃可能会或可能不会导致标度律的浮冰尺寸分布。从FSTD（海冰浓度，平均厚度，体积）得到的平均量的时间演变是复杂的，即使在简单的情景，这表明这些量，影响气候反馈，应仔细计算在气候模式。FSTD的出现与多个单独的幂律政权，如观察中所见，被认为是由于多个尺度选择过程的组合。在假设一个幂律FSTD的限制进行了仔细分析，适用于FSTD模型输出的观测方法。两个重要的误差来源，可能会导致模型偏差：一个是当观察到一个不够小的范围内的絮凝物的大小，和一个从事实上，絮凝规模的过程往往不会产生幂律行为。这两个误差来源很容易导致FSTD得出的平均量偏差大于100%，从而导致模拟海冰演变的偏差。简明语言摘要海冰是由被称为浮冰的个体碎片组成的极其复杂的镶嵌体，其大小从厘米到数十公里不等。这些浮冰的精确分布是气候模拟中的一个重要因素，因为它会影响冰、海洋和大气的共同演变。观测浮冰大小的全部范围是一项严峻的技术挑战，长期以来，海冰一直被认为具有浮冰大小的分形分布，可以用幂律来描述。我们解决这个假设是否是明智的，明确模拟浮冰如何应对气候强迫，导致浮冰融化，碰撞和断裂。我们发现，浮冰大小的分布往往不是一个幂律，并假设这种幂律行为所造成的错误可能会导致显着的错误，在如何海冰模拟。在某些情况下，幂律行为被观察到，就像在长时间的模拟中，当波浪打破彼此碰撞的浮冰时。我们强调，在做出将被纳入新气候模型的假设之前，需要更详细地观察和了解这个令人难以置信的丰富和复杂的系统。
The sub-gridscale floe size and thickness distribution (FSTD) is an emerging climate variable, playing a leading-order role in the coupling between sea ice, the ocean, and the atmosphere. The FSTD, however, is difficult to measure given the vast range of horizontal scales of individual floes, leading to the common use of power-law scaling to describe it. The evolution of a coupled mixed-layer-FSTD model of a typical marginal ice zone is explicitly simulated here, to develop a deeper understanding of how processes active at the floe scale may or may not lead to scaling laws in the floe size distribution. The time evolution of mean quantities obtained from the FSTD (sea ice concentration, mean thickness, volume) is complex even in simple scenarios, suggesting that these quantities, which affect climate feedbacks, should be carefully calculated in climate models. The emergence of FSTDs with multiple separate power-law regimes, as seen in observations, is found to be due to the combination of multiple scale-selective processes. Limitations in assuming a power-law FSTD are carefully analyzed, applying methods used in observations to FSTD model output. Two important sources of error are identified that may lead to model biases: one when observing an insufficiently small range of floe sizes, and one from the fact that floe-scale processes often do not produce power-law behavior. These two sources of error may easily lead to biases in mean quantities derived from the FSTD of greater than 100%, and therefore biases in modeled sea ice evolution.Plain Language Summary Sea ice is an incredibly complex mosaic of individual pieces, known as floes, with sizes that range from centimeters to tens of kilometers. The precise distribution of these floes is an important factor in climate simulation, as it affects how the ice, ocean, and the atmosphere evolve together. Observing the full range of floe sizes is a serious technical challenge, and sea ice has long been assumed to have a fractal distribution of floe sizes, one that can be described by a power law. We address whether this assumption is sensible by explicitly simulating how floes respond to climate forcing, leading to floes that melt, collide, and fracture. We find that the distribution of floe sizes is often not a power law, and errors caused by assuming this sort of power law behavior can lead to significant errors in how sea ice is simulated. In some cases, power-law behavior is observed, like in long simulations when waves break ice floes that are colliding with one another. We highlight the need to observe and understand this incredibly rich and complex system in greater detail before making assumptions that will be incorporated into new climate models.