Data from: Direct measurement of ant predation of weed seeds in wheat cropping

The ecosystem service of predation of weed seeds by naturally occurring seed-eating animals, including ants, in agricultural fields has been suggested to be a potentially important biocontrol option. Laboratory and field tests have found high levels of seed removal from experimentally placed seed; however, the effect of predation on naturally dispersed weed seeds is unknown. We measured the effect of invertebrate seed predators on natural weed seed dispersal and germination in a field experiment under commercial growing conditions. The two-factor, large-scale field experiment in a field used to grow wheat with conservation tillage, used an insecticide to remove soil invertebrates (ants had been reduced by 85% relative to water controls) and shallow tillage for mechanical weed removal (no effect on ants). There was one natural weed seeding event, when a wind storm blew one single incursion of Salsola australis (Chenopodiaceae) tumbleweeds across the field. We measured the number of tumbleweeds after two months and found the removal of ants resulted in a doubling of tumbleweeds: 3383 ± 513 weeds ha−1 in the insecticide treatment plots compared with 1768 ± 100 weeds ha−1 in the water control treatment plots, and 1948 ± 227 weeds ha−1 in the rest of the field. The difference in tumbleweed germination and growth was not due to growing conditions. We measured soil nutrients before the incursion and soil moisture during the weed growing period, and found that there were higher levels of nitrogenous compounds in water control plots, but no other nutrient or moisture differences, than insecticide exclusion plots. Synthesis and applications. Our results provide evidence that the ecosystem service of ant predation of naturally dispersed weed seeds limits weed abundance in commercial cereal fields in warmer climates. The fields were not managed to increase ants; alternative conservation agricultural management methods that promote agro-ecology, including low or zero tillage and low insecticide use, could increase ant abundance and thereby reduce weeds. The use of seed predators as a type of biocontrol agent will need to be integrated with other weed management methods, including herbicides and shallow tillage, although the latter may conflict with ant survival. The use of seed predators may be particularly advantageous in organic systems, or locations where economic margins are low.

农业田间中由包括蚂蚁在内的天然食籽动物对杂草种子进行捕食的生态系统服务，已被认为是一种具有潜在重要性的生物防治手段。实验室与田间试验均发现，人工放置的种子被移除的比例极高，但捕食作用对自然扩散的杂草种子的影响尚不明确。本研究在商业化种植条件下开展田间试验，评估无脊椎动物食籽者对自然扩散杂草种子的传播与萌发的影响。该试验为双因子大规模田间试验，设于采用保护性耕作的小麦田，通过施用杀虫剂去除土壤无脊椎动物（相较于清水对照组，蚂蚁数量减少了85%），并采用浅耕进行机械除草，该操作对蚂蚁数量无影响。此次试验仅发生一次天然杂草入侵事件：一场风暴将澳洲猪毛菜（Salsola australis，藜科）的风滚草单次扩散至整个试验田。我们在两个月后对风滚草数量进行了统计，结果发现去除蚂蚁的处理使风滚草数量翻倍：杀虫剂处理样地的风滚草密度为3383 ± 513 株·公顷⁻¹，清水对照组为1768 ± 100 株·公顷⁻¹，其余田间区域的密度为1948 ± 227 株·公顷⁻¹。风滚草的萌发与生长差异并非由种植条件导致。我们在杂草种子扩散前测定了土壤养分，并在杂草生长期测定了土壤湿度，结果显示清水对照组的含氮化合物水平高于杀虫剂排除样地，其余养分与湿度指标则无显著差异。综合与应用。本研究结果表明，在温暖气候下的商业化谷类农田中，蚂蚁捕食自然扩散杂草种子的生态系统服务可抑制杂草丰度。本试验田并未通过人为管理措施提升蚂蚁种群数量；而可促进农业生态的保护性耕作管理手段，包括少耕或免耕、减少杀虫剂使用等，可提升蚂蚁种群数量，进而抑制杂草生长。将食籽动物作为生物防治因子的应用需与其他杂草管理手段（包括除草剂与浅耕）相结合，不过浅耕可能会对蚂蚁的存活产生不利影响。在有机种植系统或经济利润较低的区域，利用食籽动物进行生物防治的优势尤为显著。