Biome-BGC MuSo model outputs of net ecosystem production, dry matter and water use efficiency for New Zealand under land-use scenarios

New Zealand national annual mean net ecosystem production (gC m-2 yr-1), dry matter (kg ha-1 yr-1) and water use efficiency (gross primary production divided by evapotranspriation; gC mm-1) modelled with the Biome-BGC MuSo model v6.1 (Hidy et al., 2016, 2022), for six types of landcovers: dairy pasture, sheep/beef pasture, ungrazed grasslands, exotic needleleaf forest, evergreen broadleaf forest, and shrub, over ten years 2006-2015. The "dairy" and "sheep/beef" model ecophysical parameters were calibrated against eddy covariance data from five sites in New Zealand (Bukosa et al., 2025). For all other landcovers, default parameters were used. The baseline landcover is based on the New Zealand land cover database v5.0 (LCDBv5). The four land use scenarios consist of different afforestation strategies and involve replacing grasslands with exotic needleleaf forest (pine) in different locations: 1) “blanket extended” afforestation to maximize carbon sequestration (sc1), 2) “optimized” afforestation to limit downslope transport of soil and forestry debris (sc2), 3) “optimized” afforestation to reserve high-producing grasslands for pastoral agriculture (sc3), and 4) “optimized” afforestation to preserve water quantity (sc4). Scenarios are described in detail in Mourot et al., submitted. The Biome-BGC MuSo model was run on a daily time step at a 0.01o (~1 km) grid resolution for all of NZ from 2006 to 2015. Climate inputs include daily minimum and maximum air temperature, precipitation, vapor pressure deficit, and solar radiation. These variables were obtained from the New Zealand Virtual Climate Station Network. Soil information (texture, pH, and rooting depth) comes from the Fundamental Soil Layers database, re-gridded to match the VCSN data.本数据集包含新西兰全国年度平均净生态系统生产力（net ecosystem production，单位：gC m⁻² yr⁻¹）、干物质产量（单位：kg ha⁻¹ yr⁻¹）与水分利用效率（water use efficiency，即总初级生产力（gross primary production）除以蒸散（evapotranspiration），单位：gC mm⁻¹），上述变量均通过Biome-BGC MuSo模型v6.1（Hidy等，2016、2022）模拟得到。模拟覆盖6类土地覆盖类型：奶牛牧场、绵羊/肉牛牧场、未放牧草地、外来针叶林、常绿阔叶林及灌丛，时间跨度为2006至2015年共10年。其中奶牛牧场与绵羊/肉牛牧场的模型生态物理参数依据新西兰5个站点的涡度协方差数据完成校准（Bukosa等，2025），其余土地覆盖类型均采用模型默认参数。 基准土地覆盖数据集基于新西兰土地覆盖数据库v5.0（LCDBv5）构建。本数据集包含4类土地利用情景，均采用差异化造林策略，核心操作均为将草地替换为外来针叶林（松树），具体情景如下：1）“全域扩展”造林以最大化碳固存能力（情景1，sc1）；2）“优化”造林以限制土壤与林业碎屑的下坡运移（情景2，sc2）；3）“优化”造林以保留高产草地用于畜牧农业（情景3，sc3）；4）“优化”造林以维持区域水量（情景4，sc4）。各情景的详细描述参见Mourot等已投稿的研究成果。 Biome-BGC MuSo模型以0.01°（约1公里）的网格分辨率，按日时间步长运行于2006至2015年的新西兰全域。模型所需的气候输入数据包括每日最低气温、最高气温、降水量、水汽压亏缺及太阳辐射，上述数据均源自新西兰虚拟气候站网络（New Zealand Virtual Climate Station Network）。土壤信息（质地、pH值及根深）来自基础土层数据库，并经重新网格化处理以匹配VCSN数据集。