In several regions of the world, the remoteness of potential bird hotspots and lack of trained observers have often prevented countries from effectively designing proper monitoring schemes at a national scale. For many countries, it is not known whether certain bird strongholds have been missed that should be included for more complete censuses. Such gaps at national scales, sometimes large, may be detrimental for global monitoring schemes. To address this, we used the irregular participation of Sudan to the International Waterbird Census (IWC) as a case study. We designed and tested a method based on remote‐sensing data of the country’s lowlands to detect open water bodies in order to develop predictive models of the potential distribution of waterbird abundance and diversity. To identify open water bodies and their flooding duration, we used a Modified Normalized Difference Water Index (MNDWI) derived from Landsat 8 data. Field ornithological surveys were then used as ground‐truth data to estimate the method’s accuracy. The statistical results (overall accuracy = 0.972; Kappa index = 0.93) confirmed its effectiveness. Remotely sensed water bodies and additional environmental covariates were then used to build simple habitat models of the distribution of waterbird abundance and diversity based on IWC field survey data. Of the 3119 remotely sensed clusters of open water bodies, three were predicted to host more than 10 000 waterbirds, 89 more than 1000 waterbirds and five more than 30 waterbird species. Located mainly in the southern agricultural floodplains along the main rivers, these predicted waterbird strongholds are therefore recommended for inclusion in the next IWC survey in Sudan. Our findings indicate that using remote sensing to identify open water bodies combined with simple statistical modelling is likely to be a cost‐effective solution to improve IWC sampling and to enhance both waterbird and wetland monitoring in vast under‐surveyed regions.

The basic idea to build significant attribute the uncertain objects should remove. Several theories are dealing with uncertainty, soft set theory also handles this uncertainty problem which still an open area to be explored in knowledge management. The propose techniques Known as Filtering data set which used for maintained the inferior object and we need to look at the other side of attribute reduction. The propose technique are reducing the size of object firstly, then the Hybrid reduction are executed for generating the decision extractions. These filters have reduced the size of memory without losing the characteristic of information which absolutely highly efficient. By using Filtering the inferior object of Hybrid techniques are managed. As part of this proposal, an analysis of Hybrid reduction techniques. In the conclusion part Filtering the Hybrid show better result compared to Hybrid reduction.

No abstract