景觀結構與農法對苑裡地區水稻田節肢動物生物防治效果的影響

Abstract

保育型生物防治的成效多年來受到重視，在臺灣也有越來越多的推動案例。為了瞭解苑裡地區水稻田中的保育型生物防治效果，本研究利用當地的節肢動物調查資料來分析樣區周圍景觀組成與配置以及樣區所使用的農法對於水稻田節肢動物的影響，探討各個期作與農法下天敵所發揮的生物防治效果，並提出適合的景觀管理與耕作措施，作為發展保育型生物防治之參考建議。 本研究以稻害者、掠食者、擬寄生者、中性物種以及其下的重要物種與科別作為目標節肢動物群，利用FRAGSTATS軟體來量化九個尺度下的景觀結構，藉由偏最小平方迴歸(PLSR)得到各節肢動物群所適合的景觀管理尺度以及受到景觀結構之影響。也結合了水稻耕作與氣溫資料，透過廣義估計方程式(GEE)的分析來探討農法對各節肢動物群的影響。藉由族群動態曲線下面積的計算，本研究量化了各節肢動物的數量在時間上的分佈。最後透過GEE分析，將各節肢動物的天敵與食物來源數量作為解釋變數，來探討各期作與農法下各營養階層間的交互作用與生物防治效果。 結果顯示，適合稻害者的景觀管理尺度多為半徑500公尺，適合掠食者與擬寄生者的則多為半徑250或500公尺。當植生地的形狀越複雜、嵌塊體間距離越相近，對多數稻害者與掠食者數量分別有負面與正面效果。而旱作景觀雖對多數掠食者數量有正面效果，但也可能吸引到稻害者。擬寄生者對景觀的反應與稻害者較相似，可能因食性專一，受到了稻害者的吸引。中性物種數量則受到陸域水體面積佔比與形狀發散的正面效果。在農法的部分，本研究發現相較於慣行農法，有機農法在一期雖然對各節肢動物數量沒有顯著影響，但可能減緩葉蟬科等稻害者的數量增長，在二期中有機農法則同時增加了多數稻害者與天敵的數量或數量變化率，但是否提升了生物防治效果則因天敵物種而異。結果亦顯示擬寄生者在有機田中能抑制的稻害者種類較慣行田中普遍，且能對其數量產生抑制效果，掠食者不管在哪個農法下則多僅能對稻害者的數量變化率產生抑制效果。此外蜘蛛與瓢蟲的顯著抑制效果多發生在慣行田中，可能與其廣食性以及稻害者數量有關。本研究亦發現中性物種可作為蜘蛛在水稻生長前期的替代獵物。 綜合而言，本研究認為有機農法相較於慣行是提升擬寄生者生物防治效果較適當的方式。掠食者則可以利用植生地來吸引其遷入水稻田，同時干擾稻害者的遷入。水稻田周圍的旱作也需要透過作物種類與田間操作來減少對稻害者的吸引，以及減少對掠食者的衝擊以增加掠食者的溢出。也建議能由水域環境的控制來延長中性物種在水稻田的停留時間，以增加蜘蛛在水稻生長前期的食物來源。對於以上的景觀管理，本研究建議以目標區域周圍半徑500公尺作為管理的範圍。

As the effectiveness of conservation biological control has been valued for many years, there are more and more promotion cases in Taiwan. In order to understand the effectiveness of conservation biological control in the rice fields of Yuanli area, this study used the local arthropods survey data to analyze the effects of landscape and farming practices on arthropods in the rice fields. The effectiveness of biological control under different seasons and farming practices was discussed, and the suggested landscape management and farming measures were also proposed. In this study, pests, predators, parasitoids, neutral species and the important species or families under these guilds were considered as the target arthropod groups . The FRAGSTATS software was used to quantify the landscape structures under nine scales. By using partial least squares regression (PLSR), the suitable landscape management scales for each group and their responses to landscape structures were analyzed. Combined with the data of rice cultivation and temperature, the effects of farming practices on each group were analyzed using generalized estimating equations (GEE). Through the calculation of the area under population dynamic curve, this study quantified the temporal distributions of the abundance of each group. Finally, through the GEE analysis, the abundances of natural enemies and food sources of each group were used as explanatory variables to discuss the trophic interactions and the effectiveness of biological control. The results showed that the suitable landscape management scales for most pests were 500 m radii, and for most predators and parasitoids were 250 m or 500 m radii. Also, most pests decreased with increasing shape complexity and decreasing isolation of vegetation patches, while the responses of predators to vegetation was opposite. Although the dry farmland landscape had positive effects on most predators, it may also attract pests. The responses of parasitoids to the landscape structures were similar to those of pests, which may be caused by the attraction of pests to specialist parasitoids. Neutral species were positively affected by the percentage of area and the shape complexity of the water body patches. In the analysis of effects of farming practices, it was found that although organic farming had no significant effect on any arthropod abundance, it could positively affect the abundance change rate of pests like Cicadellidae in the first season. In the second season, the abundance or the abundance change rate of most pests and natural enemies were increased by organic farming, but whether the following effectiveness of biological control was enhanced depended on the species of natural enemies. The results also showed that the pest species that parasitoids controlled in organic fields were more general than in conventional fields. While parasitoids under organic farming could suppress the abundance of pests, most predators in either farming could only suppress the abundance change rate of pests. In addition, the significant suppressing effects of spiders and ladybugs mainly occurred in conventional fields, which may be related to their general feeding habits and the pest abundance. It was also indicated that neutral species could serve as alternative prey for spiders in the early stage of rice growth. This study suggests that compared to conventional farming, organic farming is recommended to enhance the biological control effectiveness by parasitoids. Vegetation could be used to promote predators’ movement into the rice fields, and simultaneously interfere with the immigration of pests. The attraction to pests as well as the impacts on predators in the dry farmlands around rice fields should be reduced through crop selection and farming practices and thus enhance the spillover of predators. It is also suggested to adjust the water environments to extend the residence periods of neutral species so that the food sources of spiders in the early stage of rice growth could be increased. For the landscape management measures above, this study suggests considering the landscape within 500 m radius around the target fields as the management range.