Project

Little Redfish

Rallus antarcticus

Scope of the project

It is accepted that the population is undergoing a sustained decline due to habitat loss and predation by American mink.

habitat and behavior

It inhabits scattered wetlands in the desert portion of southern Patagonia. Like the rest of the coots, they are shy birds, social monogamous, territorial, of irregular seasonal migration and adapted to live in wetlands with abundant vegetation.

feeding

Insufficient data! To date there is no work that has evaluated the diet in a particular way. What we know are assumptions based on studies of other coots.

reproduction

Insufficient data! Nesting records are too scarce to describe general breeding patterns.

curiosity

After 40 years without records it was declared possibly extinct in 1992. It was only in 1998 that it was rediscovered in the Río Chico basin.

context

Small Redfish Project

Since its rediscovery in 1998 our team has participated in studies and surveys in Santa Cruz, Tierra del Fuego and Chile. In 2014 we published the progress of this work up to 2008: 18 new localities added to new information regarding habitat, seasonal movements and main threats. This work is a milestone as it established the baseline for future studies and marks the beginning of the Patagonia Program’s work with the Redfish. Later in 2014, our team publishes two new localities after five years of fieldwork (2009-2014) and reaffirms the hypothesis that American Mink is one of the causes for its supposed decline: five localities with suitable habitat are documented, all of them without Redfish but with Mink. In 2017 a bachelor thesis is defended that condenses the results of the research work developed by the team between 2015 and 2016, we also participated in the Categorization of the Birds of Argentina jointly with the Ministry of Environment and Sustainable Development, in the same we established the Redfish with the category of “Endangered” at national level considering the increasing threats (habitat loss and American Mink), scarce records and low counts. Finally, in 2019 the results of the research project developed between 2015 and 2016 are published. There, we were able to establish that wetland management (burning and cattle ranching) and the presence of mink are negatively associated with the presence of redfish, we developed an efficient sampling protocol to maximize detection during playback experiments and we suggested the control of American mink populations and the closure of certain reedbed sectors as indispensable measures to contribute to reverse the conservation status of the species.

patagonia in action program

Lines of research

ECOLOGICAL REQUIREMENTS OF REDFISH IN SANTA CRUZ WETLANDS AND ADJUSTMENT OF MONITORING PROTOCOLS.

Responsible

Andrés de Miguel

State

Completed

Target

Knowing the ecological requirements of the Redfish is important for the generation of information on how human activities affect it, a fundamental step to generate management actions that guarantee the sustainability of its populations over time. On the other hand, knowing how the detection of individuals varies according to the type of voice, time of day and season, allows adjusting monitoring protocols as a first step for the future knowledge of basic aspects such as abundance and real area of occupation. Therefore, the objective of this project was to study the ecological requirements of the Red Knot in wetlands of western Santa Cruz and to adjust monitoring protocols to encourage studies in other regions of its distribution. The specific objectives of the project were to find associations between environmental variables and site occupancy of the Redfish and, in addition, to know how the time of the season, the day and the type of voice affect the detection of individuals and estimation of site occupancy.

methodology

The work was carried out in wetlands located on the margins of some of the main rivers of north-central Santa Cruz, in the basins of the Deseado, Pinturas and Chico rivers. These wetlands, with permanent water due to winter snowfall and river overflows, are dominated by Schoenoplectus californicus reedbeds and their configuration is variable and can be continuous or fragmented in small patches, sometimes with small lagoons or open water sectors. These wetlands are important for humans, as they are generally exploited by different ranches to maximize cattle, sheep and horses. Management activities include changes to the riverbeds through canals and burning at the beginning of the season. The work was carried out throughout the breeding season of the Redfish, when five sites corresponding to different ranches were visited on at least two occasions (during and after breeding). In each one, a route was randomly defined, located at the edge of the reedbed, avoiding entering the reedbed so as not to influence behavior. Each route was formed by 10 points of approximately 50 m radius, separated by at least 100 m. Once at the sampling sites, the points were characterized (characteristics of the reedbed) and the environmental conditions of the sampling day. For the second objective, two different voices of the redfish (one calling and one territorial) were played at each point, in the morning (up to three hours after sunrise) and in the afternoon (from two hours before sunset).

results and progress

In 2017 a graduate thesis was published at the University of Buenos Aires and in 2019 these same results were published in Bird Conservation International. At the same time they presented papers at conferences and part of this information was used in the management plan for the Bosques Petrificados de Jaramillo NP, in Santa Cruz.

Artificial intelligence, population genetics and remote sensing applied to the ecology and conservation of the Redfish, a threatened and little known waterbird of Southern Patagonia.

Responsible

John M. Krapovickas

State

Active

Target

The objective of this line of research is to develop conceptual and methodological tools to design a management strategy for the conservation of the Redfish. On the one hand, we seek to develop artificial intelligence (AI) tools, and on the other, to deepen the knowledge of virtually unknown aspects of its natural history. In particular, we seek to develop an AI model capable of recognizing individuals using graphical representations of their vocalizations, and to explore its usefulness for monitoring populations and studying movements (seasonal migration and natal dispersal). In this way, we seek to develop an AI model capable of providing spatial information about Redpolls from the properties of their vocalizations. To explore its usefulness to characterize behaviors such as territoriality, mating systems, parental care, and habitat use. On the other hand, we seek to characterize the population structuring at the genetic level for the four main populations that inhabit the Gallegos and Coyle river basins, the transition between Torres del Paine (Chile) and Glaciares (Argentina) national parks, the Chico river basin and, finally, the Deseado river basin. Finally, we also studied movements (seasonal migration and natal dispersal), territoriality and habitat use by banding and instrumentation of coots with satellite transmitters.

methodology

For individual recognition we used Siamese convolutional neural networks (CNNs) trained with high quality recordings of the two vocalization types of the Redfish: "song" and "duet". For the spatial ecology study we used an array of four separate microphones. To these obtained resources we apply a model that allows, considering the different arrival times of the sound, to triangulate the position of each individual. Through captures, necessary for several of the objectives, blood samples are taken for molecular analysis of population structuring to estimate variation at the genomic level. We also use a combination of techniques including banding and remote sensing, for which captures are made outside the breeding season, using mist nets and cage traps. We use GPS-UHF transmitters, and plastic and metal bands. The transmitters store information for a year in fixed bases installed in the wetlands and allow remote downloading.

Scope area

The study is developed in the province of Santa Cruz, between latitudes 46° and 52° South. In particular, it includes the basins of the large rivers that run through the Patagonian Steppe. These rivers generate low gradient alluvial valleys where seasonal floods due to snowmelt allow the development of wet grasslands, vegetated lagoons and grasslands dominated by reeds.

partners

- Dr. Gabriel Mindlin - Dynamical Systems Laboratory, University of Buenos Aires. He guarantees the correct implementation of AI tools.
- Dr. Juan Carlos Reboreda and Dr. Bettina Mahler - Laboratory of Ecology and Animal Behavior. They will provide knowledge for behavioral analysis and technical resources for genetic analysis.

results and progress

During the year 2023 we presented the project in the call for doctoral grants of CONICET, distinguishing 27th place out of 220. The grant became effective in August 2024. In 2024 we started with preliminary tests with automatic microphones that will allow us to collect recordings necessary to train the AI models and to collect the recordings necessary for the implementation of the specific objectives.

Continue exploring