1 LOOKING BACKWARDS

One of the things that we learn from the history of science is that, with some notable exceptions beloved of philosophers of science, knowledge and understanding progress over time. Looking back, we see that understanding of the natural world has (mostly) progressed. Sometimes alternative theories have awaited experimental confirmation; sometimes a new experimental technique has led to significant theoretical advances. We hope, of course, to see some of that progression, and to make a contribution to it, over the time scale of our own careers in science. It is therefore somewhat disconcerting to have something you wrote more than 30 years ago cited (in Nearing et al., 2020) as if the comments were relevant today. Things should have changed, even in hydrology.

The context is that of the availability of the new techniques of machine learning and deep learning and their application to hydrological data. Nearing et al. (2020) suggest that in many respects not much has actually changed, since I wrote about the need for a new paradigm in hydrological modelling in 1987 (Beven, 1987). They go on to suggest that machine learning and deep learning can produce models that perform just as well, if not better, than conceptual models and process‐based hydrological models, including for catchments treated as ungauged (see also Kratzert, Klotz, Shalev et al., 2019; Kratzert, Klotz, Herrnegger, et al., 2019).

Should this be considered surprising? Not necessarily in the case of individual catchments—if there are consistent anomalies or epistemic uncertainties in catchment data that mean, for example, that water balance constraints are not well met, then a deep learning (DL) model can compensate for those anomalies in ways that a conceptual model, constrained by water balance cannot. If there are consistent anomalies between the conceptual structure of a hydrological model in a particular catchment and the nature of the hydrological processes in that catchment then again a DL model might well be able to capture that behaviour better than a deficient process description (although it is worth noting that DL models are also subject to choices in structure and multiple hidden parameters; that is what gives them flexibility in fitting the training data). Nearing et al. (2020) point out that there are techniques for incorporating conservation constraints into physically constrained DL models (see also Wang, Zhang, Chang, & Li, 2020), but given the epistemic uncertainties in water and energy balances, then this might not necessarily be advantageous in obtaining better DL predictions if, for example, the observational data do not themselves provide consistent mass and energy balance closure. Indeed, recognizing this, and how to respond to it, might already represent an advance (e.g., the discussion of Beven, 2019).

中文翻译：

---

深度学习，水文过程和地方的独特性

1向后看

我们从科学史中学到的一件事是，科学哲学家，知识和理解随着时间的推移而不断发展，但有一些明显的例外。回顾过去，我们看到对自然界的理解已经（大部分）取得了进展。有时，其他理论正在等待实验的证实。有时，新的实验技术导致了重大的理论进步。当然，我们希望在我们自己的科学事业的时间范围内看到这种进步，并为此做出贡献。因此，将您三十多年前写的东西引为引用（在Nearing等，2020中）似乎有些令人不安，好像这些评论与今天的内容有关。即使在水文学方面，情况也应该发生了变化。

背景是机器学习和深度学习新技术的可用性及其在水文数据中的应用。Nearing等。（2020）表明自从我在1987年撰写水文建模的新范式的需要以来（实际上并没有太大的改变）（Beven，1987）。他们继续建议，机器学习和深度学习所产生的模型的性能与概念模型和基于过程的水文模型相比，即使不是更好，甚至适用于未处理流域（也参见Kratzert，Klotz，Shalev等） ，2019 ; Kratzert，Klotz，Herrnegger等，2019）。

应该认为这令人惊讶吗？对于单个流域而言并不一定如此—如果流域数据中存在一致的异常或认识不确定性，例如，这意味着水平衡约束得不到很好的满足，则深度学习（DL）模型可以通过以下方式补偿这些异常：受水平衡约束的概念模型无法做到。如果特定流域的水文模型的概念结构与该流域的水文过程的性质之间存在一致的异常现象，则DL模型很可能能够比缺乏过程描述更好地捕获该行为（尽管值得一提的是，DL模型还受结构和多个隐藏参数的选择的影响；这使它们可以灵活地拟合训练数据。Nearing等。（2020年）指出，存在将守恒约束纳入物理约束DL模型的技术（另见Wang，Zhang，Chang和Li，2020年），但是鉴于水和能量平衡的认识不确定性，那么这不一定是有利的例如，如果观测数据本身不能提供一致的质量和能量平衡闭合，则可以获得更好的DL预测。确实，认识到这一点以及如何应对它可能已经代表了进步（例如，关于Beven，2019的讨论）。