Eddy-covariance and LI-710 evapotranspiration measurements from paired hazelnut orchards with and without cover crops, Willamette Valley, Oregon (2024–2025)

Dataset Description This dataset contains half-hourly eddy-covariance (EC) fluxes and paired LI-COR LI-710 evapotranspiration measurements from two adjacent hazelnut (Corylus avellana) orchard plots in the Willamette Valley, Oregon, United States. The two plots share the same soil series, irrigation system, tree age, and general management but differ in tree density (2025) and orchard-floor cover: Tower 1 is deployed in a plot with a grass ground cover (Poa annua) between tree rows, and Tower 2 is deployed at an adjacent plot kept with bare ground. Each EC tower is paired with a co-located LI-COR LI-710 mounted on a tower in an adjacent row (20 ft away) and oriented at the same direction as EC instrument, enabling direct, time and footprint-synchronous comparison of research-grade open-path eddy covariance against a low-cost, grower-oriented alternative for irrigation scheduling. Temporal coverage 2024-07-21 through 2025-10-08 (field seasons only). Records are continuous at 30-minute resolution with the exceptions documented under "Known issues" below. Instrumentation Each tower is equipped with a Campbell Scientific IRGASON integrated sonic anemometer + open-path CO₂/H₂O gas analyzer at 6 m above ground, a Hukseflux NR01 four-component net radiometer, a Campbell HygroVue 10 ambient T/RH probe, a Texas Electronics TE525 tipping-bucket rain gauge, and a soil-energy-balance suite of three Hukseflux HFP01 soil-heat-flux plates with co-located Campbell TCAV soil-temperature averaging thermocouples. Each tower is controlled by a Campbell CR1000X datalogger running the Campbell EasyFlux DL program. Paired LI-COR LI-710 evapotranspiration sensors are connected to the same loggers via SDI-12. Data products Tower{1,2}_EC_30min_2024-2025_PST.csv — 30-minute EC fluxes (LE, H, ET, Fc, u⋆, momentum), full radiative components, soil-energy terms, meteorology, and Foken-style Steady-State + Integral Turbulence Characteristics (SSITC) quality flags, in AmeriFlux TOA5 format. Tower{1,2}_LI710all_PST.csv — 30-minute LI-710 ET and ancillary variables at matching timestamps. Site and instrument metadata, variable dictionaries, and processing documentation are included alongside the data. Data processing Raw turbulence data were sampled at the IRGASON's native rate (50 Hz through 2025-05-21, 20 Hz thereafter) and processed in real time by Campbell's EasyFlux DL. The on-logger pipeline includes: sample-level QC (sonic and IRGA diagnostic flags, CO₂/H₂O signal-strength thresholds, physical-plausibility bounds), Schotanus-Nieuwstadt-De Bruin humidity correction of sonic temperature, per-period double rotation (Tanner & Thurtell, 1969), a composite frequency-response correction (Moore 1986 cospectral models + Moncrieff et al. 1997 transfer functions for path averaging and sensor separation + van Dijk 2002 sonic-T transfer function + Massman 2000 block-averaging term), WPL density correction (Webb, Pearman & Leuning, 1980), soil heat flux with TCAV/CS655 storage above the plate (EasyFlux DL Manual, Appendix H), and Foken et al. (2004) SSITC quality flags. The published 30-min CSVs are the EasyFlux DL exact output. Tower 1 additionally provides SW_IN_corrected and NETRAD_corrected, derived by substituting Tower 2's SW_IN due to intermittent NR01 pyranometer issues at Tower 1. Known caveats The Tower 1 (grass) orchard was thinned to single density (20 ft x 20 ft) during the 2024/2025 off-season, so Tower 1's 2024 growing season data represents a double-density orchard structure (10 ft x 20 ft). Tower 2 (bare) was not thinned within this dataset's coverage; its 2025/2026 off-season thinning affects only subsequent releases. Early-deployment periods (late July and August 2024) have reduced coverage and the LI-710 sensors were installed in October, 2024. The LI-710 specification cuts its flow pump at ambient temperatures below 5 °C, making winter LI-710 data unreliable. See the included docs/known_issues.md for the full list. Funding This work was supported by the Western Sustainable Agriculture Research and Education (SARE) program, grant GW24-003. How to cite Please cite the dataset using the DOI at the top of this record. A machine-readable citation block is also provided in CITATION.cff.数据集说明 本数据集包含美国俄勒冈州威拉米特谷两处相邻榛树（Corylus avellana）果园样地的半小时级涡度协方差（eddy-covariance, EC）通量数据，以及配对的LI-COR LI-710型蒸散量测量数据。两处样地土壤类型、灌溉系统、树龄与常规管理均一致，但树密度（2025年统计）与果园地表覆盖存在差异：塔1部署于树行间覆有一年生早熟禾（Poa annua）草地的样地，塔2则部署于相邻的裸地表覆盖样地。每台EC塔旁均配有一台安装在相邻树行（距塔20英尺）塔架上的LI-COR LI-710传感器，其朝向与EC仪器一致，可实现时间与足迹同步的直接对比：将研究级开放式路径涡度协方差系统，与一款面向种植者的低成本灌溉调度替代方案进行比对。 时间覆盖时段：2024年7月21日至2025年10月8日（仅包含野外监测季）。数据以30分钟分辨率连续记录，例外情况详见下文“已知问题”章节。 仪器配置：每座塔架均配备：①Campbell Scientific IRGASON型集成超声风速仪+开放式路径CO₂/H₂O气体分析仪，安装于地面以上6米处；②Hukseflux NR01型四分量净辐射计；③Campbell HygroVue 10型环境温湿度探头；④Texas Electronics TE525型翻斗式雨量计；以及一套土壤能量平衡装置，包含三台Hukseflux HFP01型土壤热通量板，以及与之配套的Campbell TCAV型土壤温度平均热电偶。每座塔架由运行Campbell EasyFlux DL程序的Campbell CR1000X型数据采集器控制。配对的LI-COR LI-710型蒸散传感器通过SDI-12协议连接至同一数据采集器。 数据产品： 1. Tower{1,2}_EC_30min_2024-2025_PST.csv：30分钟级EC通量数据（包含潜热通量LE、显热通量H、蒸散量ET、CO₂通量Fc、摩擦风速u⋆、动量通量）、完整辐射分量、土壤能量项、气象数据，以及Foken式稳态+整体湍流特征（Steady-State + Integral Turbulence Characteristics, SSITC）质量标记，格式符合AmeriFlux TOA5标准。 2. Tower{1,2}_LI710all_PST.csv：匹配时间戳的30分钟级LI-710蒸散量及辅助变量数据。 随数据一并提供样地与仪器元数据、变量词典及处理文档。 数据处理流程：原始湍流数据以IRGASON的原生采样率采集（2025年5月21日前为50 Hz，此后改为20 Hz），并通过Campbell EasyFlux DL程序实时处理。数据采集器端的处理流程包括：样本级质量控制（超声与红外气体分析仪诊断标记、CO₂/H₂O信号强度阈值、物理合理性边界检查）、Schotanus-Nieuwstadt-De Bruin声温湿度校正、逐时段双旋转法（Tanner & Thurtell, 1969）、复合频率响应校正（Moore 1986年协谱模型+Moncrieff等1997年路径平均与传感器间距传递函数+van Dijk 2002年声温传递函数+Massman 2000年块平均项）、WPL密度校正（Webb, Pearman & Leuning, 1980）、基于TCAV/CS655板上土壤储热的土壤热通量计算（参考EasyFlux DL手册附录H），以及Foken等（2004年）的SSITC质量标记。本次发布的30分钟CSV文件即为EasyFlux DL的直接输出结果。塔1数据额外提供校正后的入射短波辐射SW_IN_corrected与净辐射NETRAD_corrected，该校正通过替换为塔2的入射短波辐射数据实现，原因是塔1的NR01总辐射计存在间歇性故障。 已知局限性：2024/2025休季期间，塔1（草地覆盖样地）被疏伐至单株间距（20英尺×20英尺），因此塔1的2024年生长季数据对应双株间距果园结构（10英尺×20英尺）。塔2（裸地覆盖样地）在本数据集覆盖时段内未进行疏伐，其2025/2026休季的疏伐仅影响后续发布版本。2024年7月末至8月的早期部署阶段数据覆盖范围有限，且LI-710传感器于2024年10月才完成安装。LI-710传感器的技术规格显示，其流量泵在环境温度低于5 ℃时会停止工作，导致冬季LI-710数据不可靠。完整清单请参阅附带的docs/known_issues.md文档。 资助情况：本研究工作由西部可持续农业研究与教育（SARE）计划资助，项目编号GW24-003。 引用方式：请使用本记录顶部的DOI引用本数据集。CITATION.cff文件中也提供了机器可读的引用格式。