Versatile agricultural land spatial data of the Victoria catchment NT generated by the Victoria River Water Resource Assessment

This versatile agricultural land data is a collection of raster datasets (GeoTIFF format) used to provide a synopsis of the land suitability data of the 21 crop groups and their specific irrigation management systems and seasons in the Victoria catchment of the Northern Territory. Five datasets are in this collection. \nThe definitive versatile agricultural land dataset (Ag_Versatility_14_Crops_V.tif) was determined by identifying where the largest number of the 14 selected land management options were mapped as being suitable (i.e. suitability classes 1 to 3, refer to report cited with this metadata record). This analysis summarised the suitability of the selected land management options for each pixel, and highlights those pixels that are potentially more versatile for agricultural development because they are likely to suit a larger range of land management options and enterprises eg the score of zero represents the least versatile land, while the score of 14 represents the most versatile. The data values represent the number of land management options suitable for that pixel. The selected land management options were chosen to be relative to general potential agronomic experience and development aspirations of potential stakeholders in the catchment and were derived in consultation with the agricultural viability activity in VIWRA. These selections are presented in Table 3-26 of the published report; ' Soils and land suitability for the Victoria catchment, Northern Territory’. A technical report from the CSIRO Victoria River Water Resource Assessment to the Government of Australia. Similarly, the selection of a different representative set from the land management options would result in a different versatility map outcome.\nIn addition to the selected set of 14 land management options, versatile agricultural land is also presented using the subsets of each of the irrigation types (and rainfed cropping). In this case, the land management options were assigned to rainfed (8), furrow (17), spray (23) or trickle irrigation (10). The data values represent the number of land management options suitable for that pixel. Analytical products like these help to identify land where particular types of irrigation-related infrastructure investment may be best targeted. This data provides improved land evaluation information to identify opportunities and promote detailed investigation for a range of sustainable development options. It is important to emphasize that this is a regional-scale assessment: further data collection and detailed soil physical, chemical and nutrient analyses would be required to plan development at a scheme, enterprise or property scale. Several limitations that may have a bearing on land suitability were out of scope and not assessed as part of this activity (refer to the report), these limitations include biophysical and socio-cultural. For example these versatile agricultural land raster datasets do not include consideration of the licensing of water, flood risk, contiguous land, risk of irrigation induced secondary salinity, or land tenure and other legislative controls. Some of these may be addressed elsewhere in VIWRA eg flooding was investigated by the Earth observation remote sensing group in the surface water activity. \nThe Victoria River Water Resource Assessment provides a comprehensive overview and integrated evaluation of the feasibility of aquaculture and agriculture development in the Victoria catchment NT as well as the ecological, social and cultural (indigenous water values, rights and aspirations) impacts of development.\nLineage: These versatile agricultural land raster datasets have been generated from a range of inputs and processing steps. Following is an overview. For more information refer to the CSIRO VIWRA published report ' Soils and land suitability for the Victoria catchment, Northern Territory’. A technical report from the CSIRO Victoria River Water Resource Assessment to the Government of Australia. 1. Collated existing data (relating to: soils, climate, topography, natural resources, remotely sensed, of various formats: reports, spatial vector, spatial raster etc). 2. Selection of additional soil and land attribute site data locations by a conditioned Latin hypercube statistical sampling method applied across the covariate data space. 3. Fieldwork was carried out to collect new attribute data, soil samples for analysis and build an understanding of geomorphology and landscape processes. 4. Database analysis was performed to extract the data to specific selection criteria required for the attribute to be modelled. 5. The R statistical programming environment was used for the attribute computing. Models were built from selected input data and covariate data using predictive learning from a Random Forest approach implemented in the ranger R package. 6. Create Digital Soil Mapping (DSM) attribute raster datasets. DSM data is a geo-referenced dataset, generated from field observations and laboratory data, coupled with environmental covariate data through quantitative relationships. It applies pedometrics - the use of mathematical and statistical models that combine information from soil observations with information contained in correlated environmental variables, remote sensing images and some geophysical measurements. 7. Land management options were chosen and suitability rules created for DSM attributes. 8. Suitability rules were run to produce limitation subclass datasets using a modification on the FAO methods. 9. Final suitability data created for all land management options. 10. Companion predicted reliability data was produced from the 500 individual Random Forest attribute models created. 11. QA Quality assessment of these land suitability data was conducted by two methods. Method 1: Statistical (quantitative) assessment of the "reliability" of the spatial output data presented as a raster of the Confusion Index. Method 2: Collecting independent external validation site data combined with on-ground expert (qualitative) examination of outputs during validation field trips. A two-week validation field trip was conducted using a new validation site set which was produced by a random sampling design based on conditioned Latin Hypercube sampling. The modelled land suitability value was assessed against the actual on-ground value. These results are published in the report referenced above. 12. Select the 14 land management options for each catchment in consultation with the agricultural viability activity. 13. Calculate the versatile agricultural land datasets\n

本多功能农业土地数据集集合包含5个栅格数据集（raster datasets），格式为GeoTIFF，用于概述北领地维多利亚流域21类作物群及其特定灌溉管理系统与季相的土地适宜性数据。

基准多功能农业土地数据集（Ag_Versatility_14_Crops_V.tif）通过识别14个选定土地管理方案中被标记为适宜（即适宜等级1至3，详见本元数据记录（metadata record）所引用的报告）的最多数量的区域生成。该分析针对每个像元汇总了选定土地管理方案的适宜性，突出显示了农业开发潜力更高的像元——这些像元可适配更多类型的土地管理方案与经营主体。例如，分值0代表多功能性最低的土地，分值14则代表多功能性最高的土地。数据值代表该像元所适配的土地管理方案数量。

选定的土地管理方案基于流域内潜在利益相关方的常规农艺潜力经验与发展愿景制定，且通过与VIWRA（维多利亚河水资源评估项目）的农业可行性工作小组磋商确定。上述方案选择详见已发表报告《北领地维多利亚流域土壤与土地适宜性》（澳大利亚政府委托CSIRO开展的维多利亚河水资源评估技术报告）中的表3-26。同理，若选取不同的土地管理方案代表集，将生成不同的多功能性制图结果。

除上述14个土地管理方案集外，本数据集还针对各类灌溉类型（及雨养耕作）的子集展示了多功能农业土地信息。此时，土地管理方案被划分为雨养（8种）、沟灌（17种）、喷灌（23种）与滴灌（10种）四类。数据值同样代表该像元所适配的土地管理方案数量。此类分析产品有助于识别最适合部署特定类型灌溉相关基础设施的土地。本数据集提供了更完善的土地评估信息，可用于识别开发机遇并推动针对各类可持续发展方案的详细调研。需强调的是，本评估仅针对区域尺度：若要在项目、经营主体或地块尺度开展开发规划，还需进一步的数据采集与详细的土壤物理、化学及养分分析。

部分可能影响土地适宜性的因素未纳入本次评估范围（详见报告），包括生物物理与社会文化层面的因素。例如，本多功能农业土地栅格数据集未考虑水权许可、洪涝风险、连片土地状况、灌溉诱发次生盐渍化风险、土地权属及其他法规管控要求。其中部分内容可在VIWRA的其他模块中找到对应分析——例如，地表水工作组的地球观测遥感小组已针对洪涝开展了相关研究。

维多利亚河水资源评估项目全面概述并综合评估了北领地维多利亚流域水产养殖与农业开发的可行性，同时分析了开发活动带来的生态、社会及文化（原住民水价值、权利与愿景）影响。

数据谱系：本多功能农业土地栅格数据集通过一系列输入数据与处理步骤生成，以下为概述。如需更多信息，请参阅CSIRO委托开展的《北领地维多利亚流域土壤与土地适宜性》技术报告（澳大利亚政府委托CSIRO维多利亚河水资源评估项目成果）。
1. 整合现有数据：涵盖土壤、气候、地形、自然资源、遥感等多类数据，格式包括报告文件、空间矢量数据、空间栅格数据等。
2. 选取额外土壤与土地属性样点：通过应用于协变量数据空间的条件拉丁超立方统计抽样方法完成样点选址。
3. 开展野外工作：采集新的属性数据与土壤样本用于分析，并加深对地貌与景观过程的认知。
4. 开展数据库分析：依据属性建模所需的特定筛选标准提取数据。
5. 开展属性计算：使用R统计编程环境，通过ranger R包实现的随机森林（Random Forest）预测学习方法，基于选定的输入数据与协变量数据构建模型。
6. 生成数字土壤制图（Digital Soil Mapping, DSM）属性栅格数据集：DSM数据为地理参考数据集，通过定量关系将野外观测与实验室数据与环境协变量数据相结合，应用土壤计量学（pedometrics）方法——即结合土壤观测信息、相关环境变量信息、遥感影像与部分地球物理测量信息的数学与统计模型。
7. 选定土地管理方案并为DSM属性制定适宜性规则。
8. 运行适宜性规则，基于联合国粮食及农业组织（Food and Agriculture Organization of the United Nations, FAO）方法的修改版生成限制子类数据集。
9. 生成所有土地管理方案的最终适宜性数据。
10. 基于构建的500个独立随机森林属性模型，生成配套的预测可靠性数据。
11. 开展质量评估（Quality Assurance, QA）：采用两种方法对土地适宜性数据进行质量核查。方法1：以混淆指数（Confusion Index）栅格的形式对空间输出数据的“可靠性”开展统计（定量）评估。方法2：采集独立的外部验证样点数据，并结合验证野外考察期间的地面专家（定性）检查完成评估。本次验证野外考察为期两周，使用基于条件拉丁超立方抽样的随机抽样设计生成的新验证样点集，将建模得到的土地适宜性数值与实地实测值进行对比。上述评估结果已发表于前文引用的报告中。
12. 与农业可行性工作小组磋商，为各流域选定14个土地管理方案。
13. 计算多功能农业土地数据集。