Abstract
Spatiotemporally representative Elementary Sampling Units (ESUs) are required for capturing the temporal variations in surface spatial heterogeneity through field measurements. Since inaccessibility often coexists with heterogeneity, a cost-efficient sampling design is mandatory. We proposed a sampling strategy to generate spatiotemporally representative and cost-efficient ESUs based on the conditioned Latin hypercube sampling scheme. The proposed strategy was constrained by multi-temporal Normalized Difference Vegetation Index (NDVI) imagery, and the ESUs were limited within a sampling feasible region established based on accessibility criteria. A novel criterion based on the Overlapping Area (OA) between the NDVI frequency distribution histogram from the sampled ESUs and that from the entire study area was used to assess the sampling efficiency. A case study in Wanglang National Nature Reserve in China showed that the proposed strategy improves the spatiotemporally representativeness of sampling (mean annual OA = 74.7%) compared to the single-temporally constrained (OA = 68.7%) and the random sampling (OA = 63.1%) strategies. The introduction of the feasible region constraint significantly reduces in-situ labour-intensive characterization necessities at expenses of about 9% loss in the spatiotemporal representativeness of the sampling. Our study will support the validation activities in Wanglang experimental site providing a benchmark for locating the nodes of automatic observation systems (e.g., LAINet) which need a spatially distributed and temporally fixed sampling design.
Highlights
Accurate spatiotemporal characterization of land surface heterogeneity [1,2,3] is essential for remote sensing [4] and the land surface [5] modeling
To determine the minimum number of Elementary Sampling Units (ESUs) required to capture the spatiotemporal heterogeneity of the study area, we analyzed the variations of the mean Overlapping Area (OA) of the five dates as a function of the number of ESUs (Figure 2)
We proposed a sampling strategy to generate spatiotemporally representative and cost-efficient ESUs based on the conditioned Latin hypercube methodology
Summary
Accurate spatiotemporal characterization of land surface heterogeneity [1,2,3] is essential for remote sensing [4] and the land surface [5] modeling. The assumption of surface spatial homogeneity within the Elementary Modeling Unit (EMU) (e.g., a pixel for a remote sensing image or a grid for a land surface model) induces scaling errors [6,7]. This is especially the case for coarse spatial resolution EMU and satellite land surface products with resolutions ranging from 500 m [8] to 5 km [9]. Labour-intensive field measurement collection is usually limited by budget and time constraints In this sense, the design of efficient sampling strategies preserving the statistics of the population [13,14,15] within an affordable cost is urgently needed
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
