Abstract: Kyrgyzstan is a land of towering mountains, glaciers, rushing streams, wildflowercovered meadows, forests, snow leopards, soaring eagles, and yurt-dwelling nomads. The entire nation lies astride the Tian Shan1, Chinese for “Heavenly Mountains”, one of the world's highest mountain ranges, which is 7439 m (24,400 ft) in elevation at its highest point. The nation is the second smallest of the former Soviet Central Asian republics. In
spite of Kyrgyzstan's diverse wildlife and stunning natural beauty, the nation remains little known, and, as yet, still on the frontier of international conservation efforts. The following report is the product of 12 months of research into the state of conservation and land-use in Kyrgyzstan. This effort was funded by the Fulbright Commission of the U.S. State Department, and represents the most recent findings of the author's personal environmental journey through Inner Asia, which began in 1999. When I first started my preliminary research for this project, I was extremely surprised to learn that, even though the Tian Shan Range has tremendous ecological significance for conservation efforts in middle Asia, there wasn't a single major international conservation organization with an office in the former Soviet Central Asian republics. Even more surprising was how little awareness there is of conservation issues in the Tian Shan region amongst conservation workers in neighboring areas who are attempting to preserve similar species assemblages and ecosystems to those found in the Tian Shan. Given this lack of awareness, and the great potential for the international community to make a positive contribution towards improving the current state of biodiversity conservation in Kyrgyzstan and Central Asia, I have summarized my findings on protected areas and conservation in Kyrgyzstan and the Tian Shan of Kazakhstan, Uzbekistan, and Xinjiang in the chapters below. The report begins with some brief background information on geography and society in the Kyrgyz Republic, followed by an overview of biodiversity and the state of conservation in the nation, which at the present time closely parallels the state of conservation in the other former Soviet Central Asian republics. Part IV of the report provides a catalog of all major protected areas in Kyrgyzstan and the other Tian Shan nations, followed by a list of sites in Kyrgyzstan that are as yet unprotected but merit protection. In the appendices the reader will find fairly comprehensive species lists of flora and fauna found in the Kyrgyz Republic, including lists of mammals, birds, fish, reptiles, amphibians, trees and shrubs, wildflowers, and endemic plants. In addition, a
draft paper on the history and current practice of pastoral nomadism in Kyrgyzstan has been included in Appendix A. While the research emphasis for this study was on eastern Kyrgyzstan, over the course of the study the author did have the opportunity to make brief journeys to southern Kyrgyzstan, Uzbekistan, Kazakhstan, and Xinjiang. While falling short of being a definitive survey of protected areas of the Tian Shan, the informational review which
follows is the first attempt at bringing the details of conservation efforts throughout the entire Tian Shan Range together in one place. It is hoped that this summary of biodiversity and conservation in the Tian Shan will generate interest in the region amongst conservationists, and help increase efforts to protect this surprisingly unknown range that forms an island of meadows, rivers, lakes, and forests in the arid heart of Asia.

Abstract: Photo-identification of individual snow leopards (Panthera uncia) is the primary data source for density estimation via capture-recapture statistical methods. To identify individual snow leopards in camera trap imagery, it is necessary to match individuals from a large number of images from multiple cameras and historical catalogues, which is both time-consuming and costly. The camouflaged snow leopards also make it difficult for machine learning to classify photos, as they blend in so well with the surrounding mountain environment, rendering applicable software solutions unavailable for the species. To potentially make snow leopard individual identification available via an artificial intelligence (AI) software interface, we first trained and evaluated image classification techniques for a convolutional neural network, pose invariant embeddings (PIE) (a triplet loss network), and compared the accuracy of PIE to that of the HotSpotter algorithm (a SIFT-based algorithm). Data were acquired from a curated library of free-ranging snow leopards taken in Afghanistan between 2012 and 2019 and from captive animals in zoos in Finland, Sweden, Germany, and the United States. We discovered several flaws in the initial PIE model, such as a small amount of background matching, that was addressed, albeit likely not fixed, using background subtraction (BGS) and left-right mirroring (LR) techniques which demonstrated reasonable accuracy (Rank 1: 74% Rank-5: 92%) comparable to the Hotspotter results (Rank 1: 74% Rank 2: 84%)The PIE BGS LR model, in conjunction with Hotspotter, yielded the following results: Rank-1: 85%, Rank-5: 95%, Rank-20: 99%. In general, our findings indicate that PIE BGS LR, in conjunction with HotSpotter, can classify snow leopards more accurately than using either algorithm alone.