A sampling design for a large area forest inventory: case Tanzania

Abstract

Methods for constructing a sampling design for large area forest inventories are presented. The methods, data sets used, and the procedures are demonstrated in a real setting: constructing a sampling design for the first national forest inventory for Tanzania. The approach of the paper constructs a spatial model of forests, landscape, and land use. Sampling errors of the key parameters as well as the field measurement costs of the inventory were estimated using sampling simulation on data. Forests and land use often vary within a country or an area of interest, implying that stratified sampling is an efficient inventory design. Double sampling for stratification was taken for the statistical framework. The work was motivated by the approach used by The Food and Agriculture Organization of the United Nations (FAO) in supporting nations to establish forest inventories. The approach taken deviates significantly from the traditional FAO approaches, making it possible to calculate forest resource estimates at the subnational level without increasing the costs.

Modelling above Ground Biomass in Tanzanian Miombo Woodlands Using TanDEM-X WorldDEM and Field Data

“…Its objective is to produce information to support national and international policy processes in regard to sustainable forest management (SFM), the reduction of emissions from deforestation and forest degradation (REDD+), and greenhouse gas (GHG) emissions. The NAFORMA inventory adopts a two-phase stratified systematic cluster design with double sampling for stratification [26]. In the first phase, L-shaped clusters were selected according to a grid of 5 km × 5 km and assigned to a stratification based on: (1) plots' accessibility; (2) predicted forest growing stock; and (3) slope.…”

“…Furthermore, in the second phase the stratum-specific sampling intensities were defined using optimal allocation [27]. The distance between consecutive plots within a cluster was 250 m. Further details on the sampling design can be found in Tomppo et al [26,28]. In this study the field dataset used by Solberg et al [17] and by Naesset et al [29] increased nearly five times, thus providing the possibility to better understand the effect of the increased sample size on the AGB models.…”

“…The choice of the concentric plots was driven by the need to increase the accuracy of the measurement and sampling intensity of large trees, while also saving time. The radii of the concentric circles and respective thresholds for DHB measurements were defined subjectively according to local knowledge and practical reasons [26]. Overall, measured tree heights were on average 9.3 m and ranged from 1 to 41 m, while the DBH was 17.4 cm on average and ranged between 1 cm and 210 cm.…”

Relative Efficiency of ALS and InSAR for Biomass Estimation in a Tanzanian Rainforest

“…The plot size of 700 m 2 was chosen because it corresponds to the plot size used in the recently established national forest inventory of Tanzania [29]. The maximum plot size on each location was determined by the reach of the hypsometer, and under the most challenging conditions, distance measurement started to fail at 25 m. Thus, the maximum plot size used in the current study was 1900 m 2 .…”

Forest and Forest Change Mapping with C- and L-band SAR in Liwale, Tanzania

“…C-band SAR data is provided by the European Space Agency's Envisat A(dvanced)SAR. The project should complement the National Forestry Resources Monitoring and Assessment (NAFORMA), which is a nation-wide forest inventory program that has collected a total of more than 36000 forest plots (Tomppo, 2014). This paper presents forest mapping results from both ALOS Palsar and Envisat ASAR individually as well as combined, and yearly forest change detection from ALOS Palsar only.…”