Corn field management for wintering waterfowl on eastern Long Island, New York

The study took place in corn fields in Suffolk County, Long Island, New York, 7 February – 4 April 2018 and 7 February – 10 April 2019 (Fig.1). Fields were planted for typical production corn with 15.2 cm (6 inch) spacing among plants in rows 30.5 cm (12 inch) apart. Suffolk County contains coastal wetlands, freshwater ponds, and rural landscapes where corn fields are available to wintering waterfowl. Seasonal corn yield and wildlife abundance were determined at two corn fields in 2018 and 2019 (Cutchogue [41.023 ° N, -72.511° W] and Orient Point [41.141° N, -72.278° W]) and included another corn field in 2019 (Brookhaven [40.798° N, -72.891° W]). Corn fields were divided into three sections and marked them with flagging to identify them from a distance. The mean (± SE) corn field size was 4.08 ± 0.20 ha (Cutchogue = 3.99 ha, [0.87 ha, 1.33 ha, and 1.79 ha sections]; Orient = 4.47 ha, [1.46 ha, 1.46 ha, and 1.55 ha sections]; Brookhaven = 3.78 ha, [1.26 ha, 1.26 ha, and 1.26 ha sections]). Corn field samples were taken to obtain an index of corn availability and corn depletion rates following Barney [8]. One section in each field was chopped with a brush-hog every 2 weeks until all three sections in a field were chopped. Section of standing corn were sampled once the day before chopping and after chopping once every two weeks in 2018 and weekly in 2019. Sampling was adjusted to weekly in 2019 because some sections were depleted to zero or near zero kg/ha in &lt; 2 weeks during 2018. A random sampling design was used to distribute samples throughout the field. Main transects (<em>n</em> = 3) were established perpendicular to the field edge in each section of a field (evenly spaced 20 − 26 m apart). Each sampling period, a random number generator was used to select sampling points along each main transect. The same number of samples were taken along each main transect (<em>n</em> = 4; <em>n </em>= 12 per section). A random number generator was used to determine the left or right direction of samples to be taken off of the main transect along a perpendicular transect. A random number generator was used to determine the distance of the sampling point along the perpendicular transect (between 1 – 10 m). Corn was sampled using a 1 m <strong>× </strong>1 m quadrat at each sampling point and all corn within each quadrat was collected and placed in marked plastic bags. All individual kernels, cobs full of kernels, and cobs partial covered in kernels were included in the sample and frozen within 4 hrs of sampling. In the lab, corn was thawed, kernels were removed from cobs, and samples were dried at 60℃ until a constant mass at 48 hrs and weighed to ± 0.1 g, and reported as kg/ha. Wildlife surveys were conducted at each field 8 February – 3 April 2018 and 8 February – 9 April 2019. Morning and evening surveys were conducted, switching the time of survey at each field weekly. Morning surveys occurred 30 min before to 2 h after sunrise and evening surveys were 2 h before to 30 mins after sunset. To survey two fields on the same day, one field was surveyed in the morning and another field in the evening following weekly protocol for switching survey times. Each field was surveyed 3 times per week. Observation points were adjusted accordingly to maximize clear line of site when each section was chopped. Waterfowl flew into and landed in fields during sunrise and sunset surveys. Canada geese that were in fields at the start counts were included. This scenario reduced error in counting and identifying waterfowl to species, so 100% detection was assumed. For each field, total number of waterfowl, species composition, and other wildlife were recorded. Other wildlife included blackbirds (<em>Icteridae</em>), white-tailed deer (<em>Odocoileus virginianus</em>), and wild turkeys (<em>Meleagris gallopavo</em>).本研究于2018年2月7日至4月4日、2019年2月7日至4月10日在美国纽约州长岛萨福克县的玉米田中开展（图1）。试验田采用常规玉米生产种植模式，株距为15.2 cm（6英寸），行间距为30.5 cm（12英寸）。萨福克县拥有滨海湿地、淡水池塘与乡村景观，其玉米田可供越冬水鸟栖息。 2018年和2019年分别在两处玉米田（卡楚戈[Cutchogue，41.023°N，-72.511°W]和奥连特波因特[Orient Point，41.141°N，-72.278°W]）测定季均玉米产量与野生动物丰度，2019年新增一处玉米田（布鲁克黑文[Brookhaven，40.798°N，-72.891°W]）。每个玉米田被划分为3个区域，并用标记旗进行远距离标识。所有样田的平均（±标准误（standard error，SE））面积为4.08±0.20 ha：卡楚戈样田面积3.99 ha，分为3个区域，面积分别为0.87 ha、1.33 ha与1.79 ha；奥连特波因特样田面积4.47 ha，分为3个区域，面积分别为1.46 ha、1.46 ha与1.55 ha；布鲁克黑文样田面积3.78 ha，3个区域面积均为1.26 ha。 参考Barney[8]的方法，采集玉米田样品以获取玉米可利用量与消耗速率的相关指标。每个样田的每个区域每2周用割灌机（brush-hog）进行一次收割，直至样田全部3个区域均完成收割。2018年在收割前1天以及收割后每2周对留存的玉米区域进行一次采样，2019年则改为每周采样一次。2019年调整为每周采样的原因是，2018年部分区域的玉米在不足2周内即被消耗至0或接近0 kg/ha。 采用随机抽样设计在全田分布采样点。每个样田区域均设置3条垂直于田边的主样带（n=3），样带间距均匀为20~26 m。每个采样周期，通过随机数生成器在每条主样带上选取采样点。每条主样带上的采样点数量固定为4个，即每个区域共12个采样点。通过随机数生成器确定沿主样带的采样方向（左侧或右侧），并确定沿垂直样带的采样点距离（1~10 m之间）。 每个采样点采用1 m × 1 m的样方（quadrat）进行玉米采集，收集样方内所有玉米并装入标记好的塑料袋中。样品包含所有完整玉米粒、带粒玉米棒以及部分带粒玉米棒，并在采样后4小时内进行冷冻保存。在实验室中对样品进行解冻，将玉米粒从玉米棒上剥离，于60℃下烘干48小时至恒重，称重精度为±0.1 g，最终结果以kg/ha为单位报告。 野生动物调查于2018年2月8日至4月3日、2019年2月8日至4月9日在各试验田开展。调查分为早间与晚间时段，每周轮换各试验田的调查时段。早间调查于日出前30分钟至日出后2小时开展，晚间调查于日落前2小时至日落后30分钟开展。若当日需调查两处试验田，则按照每周轮换的调查时段安排，一处于早间调查，另一处于晚间调查。每个试验田每周开展3次调查。当某区域进行收割作业时，相应调整观测点位置以确保获得清晰的观测视野。 在日出与日落时段的调查中，会记录飞入并降落至田中的水鸟。调查起始时已在田中的加拿大黑雁需纳入统计。该设计可降低水鸟种类识别与计数误差，因此假设检测率为100%。每次调查记录水鸟总数量、物种组成以及其他野生动物种类。其他野生动物包括拟鹂科（Icteridae）鸟类、白尾鹿（Odocoileus virginianus）与野火鸡（Meleagris gallopavo）。