以注意力模塊、殘差連接建構之雨量深度學習超解析度模型

Abstract

人口的過度增長、土地的開發以及化石能源的消耗在近百年來造成地球氣候的變遷。自然災害發生的頻率也因此增加，並造成許多人類的傷亡以及產業的經濟損失。為了減緩自然的衝擊與資源的消耗，各國政府機關制定了相關政策，以減緩消耗；科學家們研發全新的、乾淨的替代能源，另一方面，氣象學家們則是藉由模型的建構，來模擬並預測這些極端事件的發生，以利人們在災害來臨之前做好準備，減少損失。其中，以水資源的影響最為深遠，它是地球中最基本也是重要的循環之一，同時也是占比最重的溫室氣體，且與人類活動息息相關。我們以台灣為例，台灣雖然年降雨平均高達2,500毫米，然而人均水資源卻是低於全球平均值。這是因為台灣的崎嶇地形特色所致，再加上季風與洋流的作用，使得降水的時空間分布不均。若能預測雨量的分布，則可訂定相關的防洪或者儲水建設，以降低災害並最大化水資源的利用，故一個準確且高解析度的預測模型一直是科學家們努力研究的方向之一。現今普遍的做法是將氣象模型的模擬資料做降尺度來提升解析度以供區域性的參考。然而這些預測模型所消耗的計算資源甚鉅，且解析度有限，很難提供疆域小且地形交互作用複雜的地區有準確的預測結果。我們提出了一個以深度學習為基礎，並結合殘差連接、注意力模塊的超解析度模型，可望提升現有的氣象模型所產出之低解析度的結果之準確性和解析度。文末，我們也比較了其他氣象降尺度的方法和其他機器學習為基礎的模型，並在四種指標(平均絕對誤差、方均根誤差、皮爾森係數、結構相似性)、定量降雨預報檢測中優於其他氣象降尺度的方法。Human activities accelerate consumption of fossil fuels and produce greenhouse gases, resulting in urgent issues today: global warming and the climate changes. These indirectly cause severe natural disasters, plenty of lives suffering and huge losses of agricultural properties.To mitigate impacts on our lands, scientists are developing renewable, reusable, and clean energies and climatologists are trying to predict the extremes. While, governments are publicizing resources saving policies for more eco-friendly society and arousing environment awareness. One of the most influencing factors is the precipitation, bringing condensed water vapor onto lands. Water resources are the most significant but basic needs in society, not only support our livings, but also economics.In Taiwan, although the average annual precipitation is up to 2,500 millimeter (mm), the water allocation for each person is lower than global average, due to the drastically geographical elevation changes and uneven distribution through the year. Thus, it is crucial to track and predict the rainfall to make the most uses ofit and to prevent the floods.However, climate models have limited resolution and require intensive computational power for local-scale uses. Therefore, we proposed a deep convolutional neural network with skip connections, attention blocks, and auxiliary data concatenation, in order to downscale the low-resolution precipitation data into high-resolution one. Eventually, we compare with other climate downscaling methods and show better performance in metrics of Mean Absolute Error (MAE), Root Mean Square Error (RMSE), Pearson Correlation, structural similarity index (SSIM), and forecast indicators.