Weather-Driven Crop Models for Integrated Pest Management

Abstract

Despite the tremendous technological advancements that have been made in recent decades, the agricultural productivity and the world food supply have been stagnated worldwide. This is partly because of the increasingly unstable weather conditions. The damages that the floods and hurricanes caused in China, Honduras and Nicaragua in 1998, are unmatched for the last century and exemplifies the problem. Though natural disasters are not be avoidable, the resulting damages can be minimized if effective weather forecasting and reporting systems are developed and utilized. A case in point is the preventive measures that were taken such as adjusting the planting dates by American farmers in anticipation of a strong El Niño in early 1998, had effectively reduced the possible damages that might otherwise occur. Taiwan’s agriculture is very productive and vitally important for social and political securities. It is important that weather resources and information are fully utilized to maximize the agricultural productivity and minimize the possible damages due to adverse weather conditions. This paper illustrates that phenological models, or weather driven plant and animal models, can be applied to predict crop developmental stages and the likelihood of disease breakouts. These models are powerful tools, which when effectively applied could increase the robustness and the sustainability of the agricultural production systems in Taiwan. 儘管近數十年來，農業技術已有驚人的進步，惟全球的農業生產和糧食的供給，卻成長緩慢；部份原因是由於不穩定之天氣型態有持續增加的趨勢。例如：1998 年大陸、宏都拉斯、尼加拉瓜所發生之洪水、颶風，遠較十九世紀者來得嚴重，此即是一明顯的例證。天然災害的發生雖無法完全地避免，但是有效的天氣預報系統若能被研發和充分加以利用的話，應可將可能發生的災害降至最低。舉例而言，在1998 年初預測當年聖嬰現象（E1 Niño）顯著，美國農民依此訊息而採取調整耕種日期的因應防範措施，有效地將潛在的災害降低。台灣的農業生產豐饒，對社會和政策之安定性具有舉足輕重的影響力。因此，當前重要的課題是如何有效地利用氣候資源和天氣資訊，以增進農業生產力，並減少潛在之天然災害的發生。本文即嘗試引進生物氣候模式，修正為適用於台灣之模式，以預測作物的生育期和病蟲害的消長情形，期望該些模式能對台灣農業生產系統的韌性和競爭力的提升有所助益。